)]}' { "log": [ { "commit": "ab5318788c6725b6d5c95aff28e63af4c35a0e2c", "tree": "fb4a81d66ed06828948e3272ebe15088d405ec1e", "parents": [ "14164b46fc994bcf82963ace00372cf808a31af1", "270750dbc18a71b23d660df110e433ff9616a2d4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 31 08:59:46 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 31 08:59:46 2014 -0800" }, "message": "Merge branch \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core debug changes from Ingo Molnar:\n \"This contains mostly kernel debugging related updates:\n\n - make hung_task detection more configurable to distros\n - add final bits for x86 UV NMI debugging, with related KGDB changes\n - update the mailing-list of MAINTAINERS entries I\u0027m involved with\"\n\n* \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n hung_task: Display every hung task warning\n sysctl: Add neg_one as a standard constraint\n x86/uv/nmi, kgdb/kdb: Fix UV NMI handler when KDB not configured\n x86/uv/nmi: Fix Sparse warnings\n kgdb/kdb: Fix no KDB config problem\n MAINTAINERS: Restore \"L: linux-kernel@vger.kernel.org\" entries\n" }, { "commit": "270750dbc18a71b23d660df110e433ff9616a2d4", "tree": "9b0400fc3713bd02d31835e8d0315cfbd7ccd912", "parents": [ "2397efb1bb17595b35f31abb40d95074ebc04f1b" ], "author": { "name": "Aaron Tomlin", "email": "atomlin@redhat.com", "time": "Mon Jan 20 17:34:13 2014 +0000" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Sat Jan 25 12:13:33 2014 +0100" }, "message": "hung_task: Display every hung task warning\n\nWhen khungtaskd detects hung tasks, it prints out\nbacktraces from a number of those tasks.\n\nLimiting the number of backtraces being printed\nout can result in the user not seeing the information\nnecessary to debug the issue. The hung_task_warnings\nsysctl controls this feature.\n\nThis patch makes it possible for hung_task_warnings\nto accept a special value to print an unlimited\nnumber of backtraces when khungtaskd detects hung\ntasks.\n\nThe special value is -1. To use this value it is\nnecessary to change types from ulong to int.\n\nSigned-off-by: Aaron Tomlin \u003catomlin@redhat.com\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: David Rientjes \u003crientjes@google.com\u003e\nCc: oleg@redhat.com\nLink: http://lkml.kernel.org/r/1390239253-24030-3-git-send-email-atomlin@redhat.com\n[ Build warning fix. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "2397efb1bb17595b35f31abb40d95074ebc04f1b", "tree": "4e8694d40ddb82a231718f9e188642fdfc966f6e", "parents": [ "64389998151214c71ba59ac893180744fd880052" ], "author": { "name": "Aaron Tomlin", "email": "atomlin@redhat.com", "time": "Mon Jan 20 17:34:12 2014 +0000" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Sat Jan 25 08:59:53 2014 +0100" }, "message": "sysctl: Add neg_one as a standard constraint\n\nAdd neg_one to the list of standard constraints - will be used by the next patch.\n\nSigned-off-by: Aaron Tomlin \u003catomlin@redhat.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: David Rientjes \u003crientjes@google.com\u003e\nCc: oleg@redhat.com\nLink: http://lkml.kernel.org/r/1390239253-24030-2-git-send-email-atomlin@redhat.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "7984754b99b6c89054edc405e9d9d35810a91d36", "tree": "9828ff56e995007158fecee07666f6a5c08f403a", "parents": [ "3b96d7db3b6dc99d207bca50037274d22e48dea5" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Thu Jan 23 15:55:59 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 23 16:37:03 2014 -0800" }, "message": "kexec: add sysctl to disable kexec_load\n\nFor general-purpose (i.e. distro) kernel builds it makes sense to build\nwith CONFIG_KEXEC to allow end users to choose what kind of things they\nwant to do with kexec. However, in the face of trying to lock down a\nsystem with such a kernel, there needs to be a way to disable kexec_load\n(much like module loading can be disabled). Without this, it is too easy\nfor the root user to modify kernel memory even when CONFIG_STRICT_DEVMEM\nand modules_disabled are set. With this change, it is still possible to\nload an image for use later, then disable kexec_load so the image (or lack\nof image) can\u0027t be altered.\n\nThe intention is for using this in environments where \"perfect\"\nenforcement is hard. Without a verified boot, along with verified\nmodules, and along with verified kexec, this is trying to give a system a\nbetter chance to defend itself (or at least grow the window of\ndiscoverability) against attack in the face of a privilege escalation.\n\nIn my mind, I consider several boot scenarios:\n\n1) Verified boot of read-only verified root fs loading fd-based\n verification of kexec images.\n2) Secure boot of writable root fs loading signed kexec images.\n3) Regular boot loading kexec (e.g. kcrash) image early and locking it.\n4) Regular boot with no control of kexec image at all.\n\n1 and 2 don\u0027t exist yet, but will soon once the verified kexec series has\nlanded. 4 is the state of things now. The gap between 2 and 4 is too\nlarge, so this change creates scenario 3, a middle-ground above 4 when 2\nand 1 are not possible for a system.\n\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nCc: Eric Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "54a43d54988a3731d644fdeb7a1d6f46b4ac64c7", "tree": "87dc618d08b23d042e3ea01b3acbc56a0358cbdb", "parents": [ "5e270e254885893f8c82ab9b91caa648af3690df" ], "author": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Thu Jan 23 15:53:13 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 23 16:36:51 2014 -0800" }, "message": "numa: add a sysctl for numa_balancing\n\nAdd a working sysctl to enable/disable automatic numa memory balancing\nat runtime.\n\nThis allows us to track down performance problems with this feature and\nis generally a good idea.\n\nThis was possible earlier through debugfs, but only with special\ndebugging options set. Also fix the boot message.\n\n[akpm@linux-foundation.org: s/sched_numa_balancing/sysctl_numa_balancing/]\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nAcked-by: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "49f0ce5f92321cdcf741e35f385669a421013cb7", "tree": "d41c5a2283b32be69840cb8722ffd8cd8c38a1b0", "parents": [ "aec6a8889a98a0cd58357cd0937a25189908f191" ], "author": { "name": "Jerome Marchand", "email": "jmarchan@redhat.com", "time": "Tue Jan 21 15:49:14 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 21 16:19:44 2014 -0800" }, "message": "mm: add overcommit_kbytes sysctl variable\n\nSome applications that run on HPC clusters are designed around the\navailability of RAM and the overcommit ratio is fine tuned to get the\nmaximum usage of memory without swapping. With growing memory, the\n1%-of-all-RAM grain provided by overcommit_ratio has become too coarse\nfor these workload (on a 2TB machine it represents no less than 20GB).\n\nThis patch adds the new overcommit_kbytes sysctl variable that allow a\nmuch finer grain.\n\n[akpm@linux-foundation.org: coding-style fixes]\n[akpm@linux-foundation.org: fix nommu build]\nSigned-off-by: Jerome Marchand \u003cjmarchan@redhat.com\u003e\nCc: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1724813d9f2c7ff702b46d3e4a4f6d9b10a8f8c2", "tree": "6df01bd6de3641b7f62440e22cf1ac8020b82ce5", "parents": [ "e4099a5e929435cd6349343f002583f29868c900" ], "author": { "name": "Peter Zijlstra", "email": "peterz@infradead.org", "time": "Tue Dec 17 12:44:49 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Jan 13 13:47:23 2014 +0100" }, "message": "sched/deadline: Remove the sysctl_sched_dl knobs\n\nRemove the deadline specific sysctls for now. The problem with them is\nthat the interaction with the exisiting rt knobs is nearly impossible\nto get right.\n\nThe current (as per before this patch) situation is that the rt and dl\nbandwidth is completely separate and we enforce rt+dl \u003c 100%. This is\nundesirable because this means that the rt default of 95% leaves us\nhardly any room, even though dl tasks are saver than rt tasks.\n\nAnother proposed solution was (a discarted patch) to have the dl\nbandwidth be a fraction of the rt bandwidth. This is highly\nconfusing imo.\n\nFurthermore neither proposal is consistent with the situation we\nactually want; which is rt tasks ran from a dl server. In which case\nthe rt bandwidth is a direct subset of dl.\n\nSo whichever way we go, the introduction of dl controls at this point\nis painful. Therefore remove them and instead share the rt budget.\n\nThis means that for now the rt knobs are used for dl admission control\nand the dl runtime is accounted against the rt runtime. I realise that\nthis isn\u0027t entirely desirable either; but whatever we do we appear to\nneed to change the interface later, so better have a small interface\nfor now.\n\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/n/tip-zpyqbqds1r0vyxtxza1e7rdc@git.kernel.org\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "332ac17ef5bfcff4766dfdfd3b4cdf10b8f8f155", "tree": "84c6663542da4310c5c555afaac88ac9b696fe4b", "parents": [ "2d3d891d3344159d5b452a645e355bbe29591e8b" ], "author": { "name": "Dario Faggioli", "email": "raistlin@linux.it", "time": "Thu Nov 07 14:43:45 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Jan 13 13:46:42 2014 +0100" }, "message": "sched/deadline: Add bandwidth management for SCHED_DEADLINE tasks\n\nIn order of deadline scheduling to be effective and useful, it is\nimportant that some method of having the allocation of the available\nCPU bandwidth to tasks and task groups under control.\nThis is usually called \"admission control\" and if it is not performed\nat all, no guarantee can be given on the actual scheduling of the\n-deadline tasks.\n\nSince when RT-throttling has been introduced each task group have a\nbandwidth associated to itself, calculated as a certain amount of\nruntime over a period. Moreover, to make it possible to manipulate\nsuch bandwidth, readable/writable controls have been added to both\nprocfs (for system wide settings) and cgroupfs (for per-group\nsettings).\n\nTherefore, the same interface is being used for controlling the\nbandwidth distrubution to -deadline tasks and task groups, i.e.,\nnew controls but with similar names, equivalent meaning and with\nthe same usage paradigm are added.\n\nHowever, more discussion is needed in order to figure out how\nwe want to manage SCHED_DEADLINE bandwidth at the task group level.\nTherefore, this patch adds a less sophisticated, but actually\nvery sensible, mechanism to ensure that a certain utilization\ncap is not overcome per each root_domain (the single rq for !SMP\nconfigurations).\n\nAnother main difference between deadline bandwidth management and\nRT-throttling is that -deadline tasks have bandwidth on their own\n(while -rt ones doesn\u0027t!), and thus we don\u0027t need an higher level\nthrottling mechanism to enforce the desired bandwidth.\n\nThis patch, therefore:\n\n - adds system wide deadline bandwidth management by means of:\n * /proc/sys/kernel/sched_dl_runtime_us,\n * /proc/sys/kernel/sched_dl_period_us,\n that determine (i.e., runtime / period) the total bandwidth\n available on each CPU of each root_domain for -deadline tasks;\n\n - couples the RT and deadline bandwidth management, i.e., enforces\n that the sum of how much bandwidth is being devoted to -rt\n -deadline tasks to stay below 100%.\n\nThis means that, for a root_domain comprising M CPUs, -deadline tasks\ncan be created until the sum of their bandwidths stay below:\n\n M * (sched_dl_runtime_us / sched_dl_period_us)\n\nIt is also possible to disable this bandwidth management logic, and\nbe thus free of oversubscribing the system up to any arbitrary level.\n\nSigned-off-by: Dario Faggioli \u003craistlin@linux.it\u003e\nSigned-off-by: Juri Lelli \u003cjuri.lelli@gmail.com\u003e\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/1383831828-15501-12-git-send-email-juri.lelli@gmail.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "1bd53a7efdc988163ec4c25f656df38dbe500632", "tree": "6bc1d2fb0b550ca3f0d86f72c2b65fbe991baf83", "parents": [ "ffe732c2430c55074bebb172d33d909c662cd0e3" ], "author": { "name": "Wanpeng Li", "email": "liwanp@linux.vnet.ibm.com", "time": "Thu Dec 12 15:23:23 2013 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Dec 17 15:24:38 2013 +0100" }, "message": "sched/numa: Drop sysctl_numa_balancing_settle_count sysctl\n\ncommit 887c290e (sched/numa: Decide whether to favour task or group weights\nbased on swap candidate relationships) drop the check against\nsysctl_numa_balancing_settle_count, this patch remove the sysctl.\n\nSigned-off-by: Wanpeng Li \u003cliwanp@linux.vnet.ibm.com\u003e\nAcked-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: David Rientjes \u003crientjes@google.com\u003e\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Naoya Horiguchi \u003cn-horiguchi@ah.jp.nec.com\u003e\nLink: http://lkml.kernel.org/r/1386833006-6600-1-git-send-email-liwanp@linux.vnet.ibm.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "5e30025a319910695f5010dc0fb53a23299da14d", "tree": "4292bcf78de221c7de1774ccf5ad0ac5a9315c26", "parents": [ "7971e23a66c94f1b9bd2d64a3e86dfbfa8c60121", "90d3839b90fe379557dae4a44735a6af78f42885" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 14 16:30:30 2013 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 14 16:30:30 2013 +0900" }, "message": "Merge branch \u0027core-locking-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core locking changes from Ingo Molnar:\n \"The biggest changes:\n\n - add lockdep support for seqcount/seqlocks structures, this\n unearthed both bugs and required extra annotation.\n\n - move the various kernel locking primitives to the new\n kernel/locking/ directory\"\n\n* \u0027core-locking-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)\n block: Use u64_stats_init() to initialize seqcounts\n locking/lockdep: Mark __lockdep_count_forward_deps() as static\n lockdep/proc: Fix lock-time avg computation\n locking/doc: Update references to kernel/mutex.c\n ipv6: Fix possible ipv6 seqlock deadlock\n cpuset: Fix potential deadlock w/ set_mems_allowed\n seqcount: Add lockdep functionality to seqcount/seqlock structures\n net: Explicitly initialize u64_stats_sync structures for lockdep\n locking: Move the percpu-rwsem code to kernel/locking/\n locking: Move the lglocks code to kernel/locking/\n locking: Move the rwsem code to kernel/locking/\n locking: Move the rtmutex code to kernel/locking/\n locking: Move the semaphore core to kernel/locking/\n locking: Move the spinlock code to kernel/locking/\n locking: Move the lockdep code to kernel/locking/\n locking: Move the mutex code to kernel/locking/\n hung_task debugging: Add tracepoint to report the hang\n x86/locking/kconfig: Update paravirt spinlock Kconfig description\n lockstat: Report avg wait and hold times\n lockdep, x86/alternatives: Drop ancient lockdep fixup message\n ...\n" }, { "commit": "7833819d2e5b2a0189f6e9d445ba03f4d7ba2656", "tree": "e6fc6a60153b0f669e3c57c203666982f00fed0e", "parents": [ "1310a5a99d900ee30b9f171146406bde0c6c2bd4" ], "author": { "name": "Chen Gang", "email": "gang.chen@asianux.com", "time": "Tue Nov 12 15:11:21 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 13 12:09:33 2013 +0900" }, "message": "kernel/sysctl.c: check return value after call proc_put_char() in __do_proc_doulongvec_minmax()\n\nNeed to check the return value of proc_put_char(), as was done in\n__do_proc_doulongvec_minmax().\n\nSigned-off-by: Chen Gang \u003cgang.chen@asianux.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "39cf275a1a18ba3c7eb9b986c5c9b35b57332798", "tree": "40b119ca9d2fbaf8128d3fa25f4c64669002b0c0", "parents": [ "ad5d69899e52792671c1aa6c7360464c7edfe09c", "e5137b50a0640009fd63a3e65c14bc6e1be8796a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 12 10:20:12 2013 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 12 10:20:12 2013 +0900" }, "message": "Merge branch \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler changes from Ingo Molnar:\n \"The main changes in this cycle are:\n\n - (much) improved CONFIG_NUMA_BALANCING support from Mel Gorman, Rik\n van Riel, Peter Zijlstra et al. Yay!\n\n - optimize preemption counter handling: merge the NEED_RESCHED flag\n into the preempt_count variable, by Peter Zijlstra.\n\n - wait.h fixes and code reorganization from Peter Zijlstra\n\n - cfs_bandwidth fixes from Ben Segall\n\n - SMP load-balancer cleanups from Peter Zijstra\n\n - idle balancer improvements from Jason Low\n\n - other fixes and cleanups\"\n\n* \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (129 commits)\n ftrace, sched: Add TRACE_FLAG_PREEMPT_RESCHED\n stop_machine: Fix race between stop_two_cpus() and stop_cpus()\n sched: Remove unnecessary iteration over sched domains to update nr_busy_cpus\n sched: Fix asymmetric scheduling for POWER7\n sched: Move completion code from core.c to completion.c\n sched: Move wait code from core.c to wait.c\n sched: Move wait.c into kernel/sched/\n sched/wait: Fix __wait_event_interruptible_lock_irq_timeout()\n sched: Avoid throttle_cfs_rq() racing with period_timer stopping\n sched: Guarantee new group-entities always have weight\n sched: Fix hrtimer_cancel()/rq-\u003elock deadlock\n sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining\n sched: Fix race on toggling cfs_bandwidth_used\n sched: Remove extra put_online_cpus() inside sched_setaffinity()\n sched/rt: Fix task_tick_rt() comment\n sched/wait: Fix build breakage\n sched/wait: Introduce prepare_to_wait_event()\n sched/wait: Add ___wait_cond_timeout() to wait_event*_timeout() too\n sched: Remove get_online_cpus() usage\n sched: Fix race in migrate_swap_stop()\n ...\n" }, { "commit": "ad5d69899e52792671c1aa6c7360464c7edfe09c", "tree": "21833c1fdab4b3cf791d4fdc86dd578e4a620514", "parents": [ "ef1417a5a6a400dbc1a2f44da716ab146a29ddc4", "caea6cf52139116e43e615d87fcbf9823e197fdf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 12 10:06:34 2013 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 12 10:06:34 2013 +0900" }, "message": "Merge branch \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf updates from Ingo Molnar:\n \"As a first remark I\u0027d like to note that the way to build perf tooling\n has been simplified and sped up, in the future it should be enough for\n you to build perf via:\n\n cd tools/perf/\n make install\n\n (ie without the -j option.) The build system will figure out the\n number of CPUs and will do a parallel build+install.\n\n The various build system inefficiencies and breakages Linus reported\n against the v3.12 pull request should now be resolved - please\n (re-)report any remaining annoyances or bugs.\n\n Main changes on the perf kernel side:\n\n * Performance optimizations:\n . perf ring-buffer code optimizations, by Peter Zijlstra\n . perf ring-buffer code optimizations, by Oleg Nesterov\n . x86 NMI call-stack processing optimizations, by Peter Zijlstra\n . perf context-switch optimizations, by Peter Zijlstra\n . perf sampling speedups, by Peter Zijlstra\n . x86 Intel PEBS processing speedups, by Peter Zijlstra\n\n * Enhanced hardware support:\n . for Intel Ivy Bridge-EP uncore PMUs, by Zheng Yan\n . for Haswell transactions, by Andi Kleen, Peter Zijlstra\n\n * Core perf events code enhancements and fixes by Oleg Nesterov:\n . for uprobes, if fork() is called with pending ret-probes\n . for uprobes platform support code\n\n * New ABI details by Andi Kleen:\n . Report x86 Haswell TSX transaction abort cost as weight\n\n Main changes on the perf tooling side (some of these tooling changes\n utilize the above kernel side changes):\n\n * \u0027perf report/top\u0027 enhancements:\n\n . Convert callchain children list to rbtree, greatly reducing the\n time taken for callchain processing, from Namhyung Kim.\n\n . Add new COMM infrastructure, further improving histogram\n processing, from Frédéric Weisbecker, one fix from Namhyung Kim.\n\n . Add /proc/kcore based live-annotation improvements, including\n build-id cache support, multi map \u0027call\u0027 instruction navigation\n fixes, kcore address validation, objdump workarounds. From\n Adrian Hunter.\n\n . Show progress on histogram collapsing, that can take a long\n time, from Namhyung Kim.\n\n . Add --max-stack option to limit callchain stack scan in \u0027top\u0027\n and \u0027report\u0027, improving callchain processing when reducing the\n stack depth is an option, from Waiman Long.\n\n . Add new option --ignore-vmlinux for perf top, from Willy\n Tarreau.\n\n * \u0027perf trace\u0027 enhancements:\n\n . \u0027perf trace\u0027 now can can use a \u0027perf probe\u0027 dynamic tracepoints\n to hook into the userspace -\u003e kernel pathname copy so that it\n can map fds to pathnames without reading /proc/pid/fd/ symlinks.\n From Arnaldo Carvalho de Melo.\n\n . Show VFS path associated with fd in live sessions, using a\n \u0027vfs_getname\u0027 \u0027perf probe\u0027 created dynamic tracepoint or by\n looking at /proc/pid/fd, from Arnaldo Carvalho de Melo.\n\n . Add \u0027trace\u0027 beautifiers for lots of syscall arguments, from\n Arnaldo Carvalho de Melo.\n\n . Implement more compact \u0027trace\u0027 output by suppressing zeroed\n args, from Arnaldo Carvalho de Melo.\n\n . Show thread COMM by default in \u0027trace\u0027, from Arnaldo Carvalho de\n Melo.\n\n . Add option to show full timestamp in \u0027trace\u0027, from David Ahern.\n\n . Add \u0027record\u0027 command in \u0027trace\u0027, to record raw_syscalls:*, from\n David Ahern.\n\n . Add summary option to dump syscall statistics in \u0027trace\u0027, from\n David Ahern.\n\n . Improve error messages in \u0027trace\u0027, providing hints about system\n configuration steps needed for using it, from Ramkumar\n Ramachandra.\n\n . \u0027perf trace\u0027 now emits hints as to why tracing is not possible,\n helping the user to setup the system to allow tracing in the\n desired permission granularity, telling if the problem is due to\n debugfs not being mounted or with not enough permission for\n !root, /proc/sys/kernel/perf_event_paranoit value, etc. From\n Arnaldo Carvalho de Melo.\n\n * \u0027perf record\u0027 enhancements:\n\n . Check maximum frequency rate for record/top, emitting better\n error messages, from Jiri Olsa.\n\n . \u0027perf record\u0027 code cleanups, from David Ahern.\n\n . Improve write_output error message in \u0027perf record\u0027, from Adrian\n Hunter.\n\n . Allow specifying B/K/M/G unit to the --mmap-pages arguments,\n from Jiri Olsa.\n\n . Fix command line callchain attribute tests to handle the new\n -g/--call-chain semantics, from Arnaldo Carvalho de Melo.\n\n * \u0027perf kvm\u0027 enhancements:\n\n . Disable live kvm command if timerfd is not supported, from David\n Ahern.\n\n . Fix detection of non-core features, from David Ahern.\n\n * \u0027perf list\u0027 enhancements:\n\n . Add usage to \u0027perf list\u0027, from David Ahern.\n\n . Show error in \u0027perf list\u0027 if tracepoints not available, from\n Pekka Enberg.\n\n * \u0027perf probe\u0027 enhancements:\n\n . Support \"$vars\" meta argument syntax for local variables,\n allowing asking for all possible variables at a given probe\n point to be collected when it hits, from Masami Hiramatsu.\n\n * \u0027perf sched\u0027 enhancements:\n\n . Address the root cause of that \u0027perf sched\u0027 stack initialization\n build slowdown, by programmatically setting a big array after\n moving the global variable back to the stack. Fix from Adrian\n Hunter.\n\n * \u0027perf script\u0027 enhancements:\n\n . Set up output options for in-stream attributes, from Adrian\n Hunter.\n\n . Print addr by default for BTS in \u0027perf script\u0027, from Adrian\n Juntmer\n\n * \u0027perf stat\u0027 enhancements:\n\n . Improved messages when doing profiling in all or a subset of\n CPUs using a workload as the session delimitator, as in:\n\n \u0027perf stat --cpu 0,2 sleep 10s\u0027\n\n from Arnaldo Carvalho de Melo.\n\n . Add units to nanosec-based counters in \u0027perf stat\u0027, from David\n Ahern.\n\n . Remove bogus info when using \u0027perf stat\u0027 -e cycles/instructions,\n from Ramkumar Ramachandra.\n\n * \u0027perf lock\u0027 enhancements:\n\n . \u0027perf lock\u0027 fixes and cleanups, from Davidlohr Bueso.\n\n * \u0027perf test\u0027 enhancements:\n\n . Fixup PERF_SAMPLE_TRANSACTION handling in sample synthesizing\n and \u0027perf test\u0027, from Adrian Hunter.\n\n . Clarify the \"sample parsing\" test entry, from Arnaldo Carvalho\n de Melo.\n\n . Consider PERF_SAMPLE_TRANSACTION in the \"sample parsing\" test,\n from Arnaldo Carvalho de Melo.\n\n . Memory leak fixes in \u0027perf test\u0027, from Felipe Pena.\n\n * \u0027perf bench\u0027 enhancements:\n\n . Change the procps visible command-name of invididual benchmark\n tests plus cleanups, from Ingo Molnar.\n\n * Generic perf tooling infrastructure/plumbing changes:\n\n . Separating data file properties from session, code\n reorganization from Jiri Olsa.\n\n . Fix version when building out of tree, as when using one of\n these:\n\n $ make help | grep perf\n perf-tar-src-pkg - Build perf-3.12.0.tar source tarball\n perf-targz-src-pkg - Build perf-3.12.0.tar.gz source tarball\n perf-tarbz2-src-pkg - Build perf-3.12.0.tar.bz2 source tarball\n perf-tarxz-src-pkg - Build perf-3.12.0.tar.xz source tarball\n $\n\n from David Ahern.\n\n . Enhance option parse error message, showing just the help lines\n of the options affected, from Namhyung Kim.\n\n . libtraceevent updates from upstream trace-cmd repo, from Steven\n Rostedt.\n\n . Always use perf_evsel__set_sample_bit to set sample_type, from\n Adrian Hunter.\n\n . Memory and mmap leak fixes from Chenggang Qin.\n\n . Assorted build fixes for from David Ahern and Jiri Olsa.\n\n . Speed up and prettify the build system, from Ingo Molnar.\n\n . Implement addr2line directly using libbfd, from Roberto Vitillo.\n\n . Separate the GTK support in a separate libperf-gtk.so DSO, that\n is only loaded when --gtk is specified, from Namhyung Kim.\n\n . perf bash completion fixes and improvements from Ramkumar\n Ramachandra.\n\n . Support for Openembedded/Yocto -dbg packages, from Ricardo\n Ribalda Delgado.\n\n And lots and lots of other fixes and code reorganizations that did not\n make it into the list, see the shortlog, diffstat and the Git log for\n details!\"\n\n* \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (300 commits)\n uprobes: Fix the memory out of bound overwrite in copy_insn()\n uprobes: Fix the wrong usage of current-\u003eutask in uprobe_copy_process()\n perf tools: Remove unneeded include\n perf record: Remove post_processing_offset variable\n perf record: Remove advance_output function\n perf record: Refactor feature handling into a separate function\n perf trace: Don\u0027t relookup fields by name in each sample\n perf tools: Fix version when building out of tree\n perf evsel: Ditch evsel-\u003ehandler.data field\n uprobes: Export write_opcode() as uprobe_write_opcode()\n uprobes: Introduce arch_uprobe-\u003eixol\n uprobes: Kill module_init() and module_exit()\n uprobes: Move function declarations out of arch\n perf/x86/intel: Add Ivy Bridge-EP uncore IRP box support\n perf/x86/intel/uncore: Add filter support for IvyBridge-EP QPI boxes\n perf: Factor out strncpy() in perf_event_mmap_event()\n tools/perf: Add required memory barriers\n perf: Fix arch_perf_out_copy_user default\n perf: Update a stale comment\n perf: Optimize perf_output_begin() -- address calculation\n ...\n" }, { "commit": "c90423d1de12fbeaf0c898e1db0e962de347302b", "tree": "8c7a32b37e74155324ae2b556fcc42718ccf29a3", "parents": [ "ecf1f014325ba60f4df35edae1a357c67c5d4eb1", "b8a216269ec0ce2e961d32e6d640d7010b8a818e" ], "author": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Nov 06 07:50:37 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Nov 06 07:50:37 2013 +0100" }, "message": "Merge branch \u0027sched/core\u0027 into core/locking, to prepare the kernel/locking/ file move\n\nConflicts:\n\tkernel/Makefile\n\nThere are conflicts in kernel/Makefile due to file moving in the\nscheduler tree - resolve them.\n\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "8eaede49dfdc1ff1d727f9c913665b8009945191", "tree": "094c592c127142c4f01766fb82e3a7973c87ca1c", "parents": [ "e8b5cbb041130ef297c90f8af2d3d45dfb9e6d15" ], "author": { "name": "Ben Hutchings", "email": "ben@decadent.org.uk", "time": "Mon Oct 07 01:05:46 2013 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Wed Oct 16 13:01:44 2013 -0700" }, "message": "sysrq: Allow magic SysRq key functions to be disabled through Kconfig\n\nTurn the initial value of sysctl kernel.sysrq (SYSRQ_DEFAULT_ENABLE)\ninto a Kconfig variable.\n\nOriginal version by Bastian Blank \u003cwaldi@debian.org\u003e.\n\nSigned-off-by: Ben Hutchings \u003cben@decadent.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "de1c9ce6f07fec0381a39a9d0b379ea35aa1167f", "tree": "d96bf1a2b25dfa84d3fe5f6fe00fb780800e3ef3", "parents": [ "1e3646ffc64b232cb14a5ef01d7b98997c1b73f9" ], "author": { "name": "Rik van Riel", "email": "riel@redhat.com", "time": "Mon Oct 07 11:29:39 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Oct 09 14:48:21 2013 +0200" }, "message": "sched/numa: Skip some page migrations after a shared fault\n\nShared faults can lead to lots of unnecessary page migrations,\nslowing down the system, and causing private faults to hit the\nper-pgdat migration ratelimit.\n\nThis patch adds sysctl numa_balancing_migrate_deferred, which specifies\nhow many shared page migrations to skip unconditionally, after each page\nmigration that is skipped because it is a shared fault.\n\nThis reduces the number of page migrations back and forth in\nshared fault situations. It also gives a strong preference to\nthe tasks that are already running where most of the memory is,\nand to moving the other tasks to near the memory.\n\nTesting this with a much higher scan rate than the default\nstill seems to result in fewer page migrations than before.\n\nMemory seems to be somewhat better consolidated than previously,\nwith multi-instance specjbb runs on a 4 node system.\n\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/1381141781-10992-62-git-send-email-mgorman@suse.de\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "930aa174fcc8b0efaad102fd80f677b92f35eaa2", "tree": "7746bbcf350f3ce305d9e55435f7a5e3c41b9c8e", "parents": [ "04bb2f9475054298f0c67a89ca92cade42d3fe5e" ], "author": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Mon Oct 07 11:29:37 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Oct 09 14:48:18 2013 +0200" }, "message": "sched/numa: Remove the numa_balancing_scan_period_reset sysctl\n\nWith scan rate adaptions based on whether the workload has properly\nconverged or not there should be no need for the scan period reset\nhammer. Get rid of it.\n\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/1381141781-10992-60-git-send-email-mgorman@suse.de\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "3a7053b3224f4a8b0e8184166190076593621617", "tree": "dfe404bfbc1306fccbc00f2177becf1482504e45", "parents": [ "745d61476ddb737aad3495fa6d9a8f8c2ee59f86" ], "author": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Mon Oct 07 11:29:00 2013 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Oct 09 12:40:26 2013 +0200" }, "message": "sched/numa: Favour moving tasks towards the preferred node\n\nThis patch favours moving tasks towards NUMA node that recorded a higher\nnumber of NUMA faults during active load balancing. Ideally this is\nself-reinforcing as the longer the task runs on that node, the more faults\nit should incur causing task_numa_placement to keep the task running on that\nnode. In reality a big weakness is that the nodes CPUs can be overloaded\nand it would be more efficient to queue tasks on an idle node and migrate\nto the new node. This would require additional smarts in the balancer so\nfor now the balancer will simply prefer to place the task on the preferred\nnode for a PTE scans which is controlled by the numa_balancing_settle_count\nsysctl. Once the settle_count number of scans has complete the schedule\nis free to place the task on an alternative node if the load is imbalanced.\n\n[srikar@linux.vnet.ibm.com: Fixed statistics]\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\n[ Tunable and use higher faults instead of preferred. ]\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/1381141781-10992-23-git-send-email-mgorman@suse.de\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "723478c8a471403c53cf144999701f6e0c4bbd11", "tree": "d89b59d4318eb5e37082f90880ac20b18f81a020", "parents": [ "354cc40e3b0981c38ba2ce2964954480e6c03c37" ], "author": { "name": "Knut Petersen", "email": "Knut_Petersen@t-online.de", "time": "Wed Sep 25 14:29:37 2013 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Oct 04 10:06:07 2013 +0200" }, "message": "perf: Enforce 1 as lower limit for perf_event_max_sample_rate\n\n/proc/sys/kernel/perf_event_max_sample_rate will accept\nnegative values as well as 0.\n\nNegative values are unreasonable, and 0 causes a\ndivide by zero exception in perf_proc_update_handler.\n\nThis patch enforces a lower limit of 1.\n\nSigned-off-by: Knut Petersen \u003cKnut_Petersen@t-online.de\u003e\nSigned-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/5242DB0C.4070005@t-online.de\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "cd64647f043e3fd3569bcf068f47f030198ff93a", "tree": "c4ff621340758e4dde72ddb8c3a63d32604c0fbb", "parents": [ "d8524ae9d6f492a9c6db9f4d89c5f9b8782fa2d5" ], "author": { "name": "Li Zefan", "email": "lizefan@huawei.com", "time": "Mon Sep 23 16:43:58 2013 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Sep 23 11:10:49 2013 +0200" }, "message": "hung_task: Change sysctl_hung_task_check_count to \u0027int\u0027\n\nAs \u0027sysctl_hung_task_check_count\u0027 is \u0027unsigned long\u0027 when this\nvalue is assigned to max_count in check_hung_uninterruptible_tasks(),\nit\u0027s truncated to \u0027int\u0027 type.\n\nThis causes a minor artifact: if we write 2^32 to sysctl.hung_task_check_count,\nhung task detection will be effectively disabled.\n\nWith this fix, it will still truncate the user input to 32 bits, but\nreading sysctl.hung_task_check_count reflects the actual truncated value.\n\nSigned-off-by: Li Zefan \u003clizefan@huawei.com\u003e\nAcked-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nLink: http://lkml.kernel.org/r/523FFF4E.9050401@huawei.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "26935fb06ee88f1188789807687c03041f3c70d9", "tree": "381c487716540b52348d78bee6555f8fa61d77ef", "parents": [ "3cc69b638e11bfda5d013c2b75b60934aa0e88a1", "bf2ba3bc185269eca274b458aac46ba1ad7c1121" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 15:01:38 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 15:01:38 2013 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs pile 4 from Al Viro:\n \"list_lru pile, mostly\"\n\nThis came out of Andrew\u0027s pile, Al ended up doing the merge work so that\nAndrew didn\u0027t have to.\n\nAdditionally, a few fixes.\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (42 commits)\n super: fix for destroy lrus\n list_lru: dynamically adjust node arrays\n shrinker: Kill old -\u003eshrink API.\n shrinker: convert remaining shrinkers to count/scan API\n staging/lustre/libcfs: cleanup linux-mem.h\n staging/lustre/ptlrpc: convert to new shrinker API\n staging/lustre/obdclass: convert lu_object shrinker to count/scan API\n staging/lustre/ldlm: convert to shrinkers to count/scan API\n hugepage: convert huge zero page shrinker to new shrinker API\n i915: bail out earlier when shrinker cannot acquire mutex\n drivers: convert shrinkers to new count/scan API\n fs: convert fs shrinkers to new scan/count API\n xfs: fix dquot isolation hang\n xfs-convert-dquot-cache-lru-to-list_lru-fix\n xfs: convert dquot cache lru to list_lru\n xfs: rework buffer dispose list tracking\n xfs-convert-buftarg-lru-to-generic-code-fix\n xfs: convert buftarg LRU to generic code\n fs: convert inode and dentry shrinking to be node aware\n vmscan: per-node deferred work\n ...\n" }, { "commit": "86cdb465cf3a9d81058b517af05074157fa9dcdd", "tree": "47b1dcb170c81cd7afd0341123dcd39c8623385d", "parents": [ "83467efbdb7948146581a56cbd683a22a0684bbb" ], "author": { "name": "Naoya Horiguchi", "email": "n-horiguchi@ah.jp.nec.com", "time": "Wed Sep 11 14:22:13 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 11 15:57:49 2013 -0700" }, "message": "mm: prepare to remove /proc/sys/vm/hugepages_treat_as_movable\n\nNow hugepage migration is enabled, although restricted on pmd-based\nhugepages for now (due to lack of testing.) So we should allocate\nmigratable hugepages from ZONE_MOVABLE if possible.\n\nThis patch makes GFP flags in hugepage allocation dependent on migration\nsupport, not only the value of hugepages_treat_as_movable. It provides no\nchange on the behavior for architectures which do not support hugepage\nmigration,\n\nSigned-off-by: Naoya Horiguchi \u003cn-horiguchi@ah.jp.nec.com\u003e\nAcked-by: Andi Kleen \u003cak@linux.intel.com\u003e\nReviewed-by: Wanpeng Li \u003cliwanp@linux.vnet.ibm.com\u003e\nCc: Hillf Danton \u003cdhillf@gmail.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Michal Hocko \u003cmhocko@suse.cz\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: \"Aneesh Kumar K.V\" \u003caneesh.kumar@linux.vnet.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3942c07ccf98e66b8893f396dca98f5b076f905f", "tree": "063ec7aa542d9fa812482c02e2436205fe6a9e8e", "parents": [ "da5338c7498556b760871661ffecb053cc6f708f" ], "author": { "name": "Glauber Costa", "email": "glommer@openvz.org", "time": "Wed Aug 28 10:17:53 2013 +1000" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Sep 10 18:56:29 2013 -0400" }, "message": "fs: bump inode and dentry counters to long\n\nThis series reworks our current object cache shrinking infrastructure in\ntwo main ways:\n\n * Noticing that a lot of users copy and paste their own version of LRU\n lists for objects, we put some effort in providing a generic version.\n It is modeled after the filesystem users: dentries, inodes, and xfs\n (for various tasks), but we expect that other users could benefit in\n the near future with little or no modification. Let us know if you\n have any issues.\n\n * The underlying list_lru being proposed automatically and\n transparently keeps the elements in per-node lists, and is able to\n manipulate the node lists individually. Given this infrastructure, we\n are able to modify the up-to-now hammer called shrink_slab to proceed\n with node-reclaim instead of always searching memory from all over like\n it has been doing.\n\nPer-node lru lists are also expected to lead to less contention in the lru\nlocks on multi-node scans, since we are now no longer fighting for a\nglobal lock. The locks usually disappear from the profilers with this\nchange.\n\nAlthough we have no official benchmarks for this version - be our guest to\nindependently evaluate this - earlier versions of this series were\nperformance tested (details at\nhttp://permalink.gmane.org/gmane.linux.kernel.mm/100537) yielding no\nvisible performance regressions while yielding a better qualitative\nbehavior in NUMA machines.\n\nWith this infrastructure in place, we can use the list_lru entry point to\nprovide memcg isolation and per-memcg targeted reclaim. Historically,\nthose two pieces of work have been posted together. This version presents\nonly the infrastructure work, deferring the memcg work for a later time,\nso we can focus on getting this part tested. You can see more about the\nhistory of such work at http://lwn.net/Articles/552769/\n\nDave Chinner (18):\n dcache: convert dentry_stat.nr_unused to per-cpu counters\n dentry: move to per-sb LRU locks\n dcache: remove dentries from LRU before putting on dispose list\n mm: new shrinker API\n shrinker: convert superblock shrinkers to new API\n list: add a new LRU list type\n inode: convert inode lru list to generic lru list code.\n dcache: convert to use new lru list infrastructure\n list_lru: per-node list infrastructure\n shrinker: add node awareness\n fs: convert inode and dentry shrinking to be node aware\n xfs: convert buftarg LRU to generic code\n xfs: rework buffer dispose list tracking\n xfs: convert dquot cache lru to list_lru\n fs: convert fs shrinkers to new scan/count API\n drivers: convert shrinkers to new count/scan API\n shrinker: convert remaining shrinkers to count/scan API\n shrinker: Kill old -\u003eshrink API.\n\nGlauber Costa (7):\n fs: bump inode and dentry counters to long\n super: fix calculation of shrinkable objects for small numbers\n list_lru: per-node API\n vmscan: per-node deferred work\n i915: bail out earlier when shrinker cannot acquire mutex\n hugepage: convert huge zero page shrinker to new shrinker API\n list_lru: dynamically adjust node arrays\n\nThis patch:\n\nThere are situations in very large machines in which we can have a large\nquantity of dirty inodes, unused dentries, etc. This is particularly true\nwhen umounting a filesystem, where eventually since every live object will\neventually be discarded.\n\nDave Chinner reported a problem with this while experimenting with the\nshrinker revamp patchset. So we believe it is time for a change. This\npatch just moves int to longs. Machines where it matters should have a\nbig long anyway.\n\nSigned-off-by: Glauber Costa \u003cglommer@openvz.org\u003e\nCc: Dave Chinner \u003cdchinner@redhat.com\u003e\nCc: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Artem Bityutskiy \u003cartem.bityutskiy@linux.intel.com\u003e\nCc: Arve Hjønnevåg \u003carve@android.com\u003e\nCc: Carlos Maiolino \u003ccmaiolino@redhat.com\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nCc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nCc: Dave Chinner \u003cdchinner@redhat.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Gleb Natapov \u003cgleb@redhat.com\u003e\nCc: Greg Thelen \u003cgthelen@google.com\u003e\nCc: J. Bruce Fields \u003cbfields@redhat.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nCc: Jerome Glisse \u003cjglisse@redhat.com\u003e\nCc: John Stultz \u003cjohn.stultz@linaro.org\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Kent Overstreet \u003ckoverstreet@google.com\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nCc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nCc: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "d738ce8fdc05ebf5b1475f8ae26d908c8c50970b", "tree": "0ed3038b6f7950d826ca0bc24a7e98f033f88b44", "parents": [ "555445cd11803c6bc93b2be31968f3949ef7708b" ], "author": { "name": "Francesco Fusco", "email": "ffusco@redhat.com", "time": "Wed Jul 24 10:39:07 2013 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jul 26 14:22:10 2013 -0700" }, "message": "sysctl: range checking in do_proc_dointvec_ms_jiffies_conv\n\nWhen (integer) sysctl values are expressed in ms and have to be\nrepresented internally as jiffies. The msecs_to_jiffies function\nreturns an unsigned long, which gets assigned to the integer.\nThis patch prevents the value to be assigned if bigger than\nINT_MAX, done in a similar way as in cba9f3 (\"Range checking in\ndo_proc_dointvec_(userhz_)jiffies_conv\").\n\nSigned-off-by: Francesco Fusco \u003cffusco@redhat.com\u003e\nCC: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCC: linux-kernel@vger.kernel.org\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f2006e27396f55276f24434f56e208d86e7f9908", "tree": "71896db916d33888b4286f80117d3cac0da40e6d", "parents": [ "e399eb56a6110e13f97e644658648602e2b08de7", "9903883f1dd6e86f286b7bfa6e4b423f98c1cd9e" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Jul 12 12:34:42 2013 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Jul 12 12:34:42 2013 +0200" }, "message": "Merge branch \u0027linus\u0027 into timers/urgent\n\nGet upstream changes so we can apply fixes against them\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "c72bb316916b1a6cf35e1d5238566ef27b0b7f80", "tree": "1bd7bb147302abf907bba9fb83cf74b4a5b6ef0d", "parents": [ "6d128e1e72bf082542e85f72e6b7ddd704193588", "dcc302232c1f9b3ca16f6b8ee190eb0b1a8a0da3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 11 09:02:09 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 11 09:02:09 2013 -0700" }, "message": "Merge tag \u0027trace-3.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing changes from Steven Rostedt:\n \"The majority of the changes here are cleanups for the large changes\n that were added to 3.10, which includes several bug fixes that have\n been marked for stable.\n\n As for new features, there were a few, but nothing to write to LWN\n about. These include:\n\n New function trigger called \"dump\" and \"cpudump\" that will cause\n ftrace to dump its buffer to the console when the function is called.\n The difference between \"dump\" and \"cpudump\" is that \"dump\" will dump\n the entire contents of the ftrace buffer, where as \"cpudump\" will only\n dump the contents of the ftrace buffer for the CPU that called the\n function.\n\n Another small enhancement is a new sysctl switch called\n \"traceoff_on_warning\" which, when enabled, will disable tracing if any\n WARN_ON() is triggered. This is useful if you want to debug what\n caused a warning and do not want to risk losing your trace data by the\n ring buffer overwriting the data before you can disable it. There\u0027s\n also a kernel command line option that will make this enabled at boot\n up called the same thing\"\n\n* tag \u0027trace-3.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (34 commits)\n tracing: Make tracing_open_generic_{tr,tc}() static\n tracing: Remove ftrace() function\n tracing: Remove TRACE_EVENT_TYPE enum definition\n tracing: Make tracer_tracing_{off,on,is_on}() static\n tracing: Fix irqs-off tag display in syscall tracing\n uprobes: Fix return value in error handling path\n tracing: Fix race between deleting buffer and setting events\n tracing: Add trace_array_get/put() to event handling\n tracing: Get trace_array ref counts when accessing trace files\n tracing: Add trace_array_get/put() to handle instance refs better\n tracing: Protect ftrace_trace_arrays list in trace_events.c\n tracing: Make trace_marker use the correct per-instance buffer\n ftrace: Do not run selftest if command line parameter is set\n tracing/kprobes: Don\u0027t pass addr\u003dip to perf_trace_buf_submit()\n tracing: Use flag buffer_disabled for irqsoff tracer\n tracing/kprobes: Turn trace_probe-\u003efiles into list_head\n tracing: Fix disabling of soft disable\n tracing: Add missing syscall_metadata comment\n tracing: Simplify code for showing of soft disabled flag\n tracing/kprobes: Kill probe_enable_lock\n ...\n" }, { "commit": "e399eb56a6110e13f97e644658648602e2b08de7", "tree": "edcd90e3134a5409339794f43d5dd4453cd72586", "parents": [ "7c6809ff2bd63d4c97ce9e0b94d39d5180842c48", "5b8621a68fdcd2baf1d3b413726f913a5254d46a" ], "author": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Jul 10 10:43:25 2013 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Wed Jul 10 10:43:25 2013 +0200" }, "message": "Merge branch \u0027timers/core\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks into timers/urgent\n\nPull nohz updates/fixes from Frederic Weisbecker:\n\n\u0027 Note that \"watchdog: Boot-disable by default on full dynticks\" is a temporary\n solution to solve the issue with the watchdog that prevents the tick from\n stopping. This is to make sure that 3.11 doesn\u0027t have that problem as several\n people complained about it.\n\n A proper and longer term solution has been proposed by Peterz:\n\n http://lkml.kernel.org/r/20130618103632.GO3204@twins.programming.kicks-ass.net\n\u0027\n\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "14c63f17b1fde5a575a28e96547a22b451c71fb5", "tree": "781f7327f4341a3d27197e88994b1859e9b51722", "parents": [ "2ab00456ea8a0d79acb1390659b98416111880b2" ], "author": { "name": "Dave Hansen", "email": "dave.hansen@linux.intel.com", "time": "Fri Jun 21 08:51:36 2013 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Sun Jun 23 11:52:57 2013 +0200" }, "message": "perf: Drop sample rate when sampling is too slow\n\nThis patch keeps track of how long perf\u0027s NMI handler is taking,\nand also calculates how many samples perf can take a second. If\nthe sample length times the expected max number of samples\nexceeds a configurable threshold, it drops the sample rate.\n\nThis way, we don\u0027t have a runaway sampling process eating up the\nCPU.\n\nThis patch can tend to drop the sample rate down to level where\nperf doesn\u0027t work very well. *BUT* the alternative is that my\nsystem hangs because it spends all of its time handling NMIs.\n\nI\u0027ll take a busted performance tool over an entire system that\u0027s\nbusted and undebuggable any day.\n\nBTW, my suspicion is that there\u0027s still an underlying bug here.\nUsing the HPET instead of the TSC is definitely a contributing\nfactor, but I suspect there are some other things going on.\nBut, I can\u0027t go dig down on a bug like that with my machine\nhanging all the time.\n\nSigned-off-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: paulus@samba.org\nCc: acme@ghostprotocols.net\nCc: Dave Hansen \u003cdave@sr71.net\u003e\n[ Prettified it a bit. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "3c00ea82c724fab0b98f15428a804cb45eb9ad38", "tree": "3e380db50cf9bc5d2962c25cc80707bfd9fc834c", "parents": [ "b8900bc0217fac8e68085997bee2f05e6db931a2" ], "author": { "name": "Frederic Weisbecker", "email": "fweisbec@gmail.com", "time": "Sun May 19 20:45:15 2013 +0200" }, "committer": { "name": "Frederic Weisbecker", "email": "fweisbec@gmail.com", "time": "Thu Jun 20 15:41:18 2013 +0200" }, "message": "watchdog: Rename confusing state variable\n\nWe have two very conflicting state variable names in the\nwatchdog:\n\n* watchdog_enabled: This one reflects the user interface. It\u0027s\nset to 1 by default and can be overriden with boot options\nor sysctl/procfs interface.\n\n* watchdog_disabled: This is the internal toggle state that\ntells if watchdog threads, timers and NMI events are currently\nrunning or not. This state mostly depends on the user settings.\nIt\u0027s a convenient state latch.\n\nNow we really need to find clearer names because those\nare just too confusing to encourage deep review.\n\nwatchdog_enabled now becomes watchdog_user_enabled to reflect\nits purpose as an interface.\n\nwatchdog_disabled becomes watchdog_running to suggest its\nrole as a pure internal state.\n\nSigned-off-by: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Srivatsa S. Bhat \u003csrivatsa.bhat@linux.vnet.ibm.com\u003e\nCc: Anish Singh \u003canish198519851985@gmail.com\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nCc: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Borislav Petkov \u003cbp@alien8.de\u003e\nCc: Li Zhong \u003czhong@linux.vnet.ibm.com\u003e\nCc: Don Zickus \u003cdzickus@redhat.com\u003e\n" }, { "commit": "de7edd31457b626e54a0b2a7e8ff4d65492f01ad", "tree": "fc98cad3155f6cded6192709c4cb496dd17282db", "parents": [ "c3e13c7c0605677a2c94957b39157f4501cea9a8" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Jun 14 16:21:43 2013 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Jun 19 23:32:07 2013 -0400" }, "message": "tracing: Disable tracing on warning\n\nAdd a traceoff_on_warning option in both the kernel command line as well\nas a sysctl option. When set, any WARN*() function that is hit will cause\nthe tracing_on variable to be cleared, which disables writing to the\nring buffer.\n\nThis is useful especially when tracing a bug with function tracing. When\na warning is hit, the print caused by the warning can flood the trace with\nthe functions that producing the output for the warning. This can make the\nresulting trace useless by either hiding where the bug happened, or worse,\nby overflowing the buffer and losing the trace of the bug totally.\n\nAcked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "a6572f84c5b135d9b6df279ed3c8de028bd1edd9", "tree": "8555a23db262d55512bf1fc365c5be8ede427f10", "parents": [ "1b45adcd9a503428e6de6b39bc6892d86c9c1d41" ], "author": { "name": "Li Zefan", "email": "lizefan@huawei.com", "time": "Fri May 17 10:31:04 2013 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue May 28 11:28:18 2013 +0200" }, "message": "watchdog: Disallow setting watchdog_thresh to -1\n\nIn old kernels, it\u0027s allowed to set softlockup_thresh to -1 or 0\nto disable softlockup detection. However watchdog_thresh only\nuses 0 to disable detection, and setting it to -1 just froze my\nbox and nothing I can do but reboot.\n\nSigned-off-by: Li Zefan \u003clizefan@huawei.com\u003e\nAcked-by: Don Zickus \u003cdzickus@redhat.com\u003e\nLink: http://lkml.kernel.org/r/51959668.9040106@huawei.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "4eeab4f5580d11bffedc697684b91b0bca0d5009", "tree": "4e9a0c010d34e786df52225039a17aa38e9adf17", "parents": [ "c9b1d0981fcce3d9976d7b7a56e4e0503bc610dd" ], "author": { "name": "Andrew Shewmaker", "email": "agshew@gmail.com", "time": "Mon Apr 29 15:08:11 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 29 15:54:36 2013 -0700" }, "message": "mm: replace hardcoded 3% with admin_reserve_pages knob\n\nAdd an admin_reserve_kbytes knob to allow admins to change the hardcoded\nmemory reserve to something other than 3%, which may be multiple\ngigabytes on large memory systems. Only about 8MB is necessary to\nenable recovery in the default mode, and only a few hundred MB are\nrequired even when overcommit is disabled.\n\nThis affects OVERCOMMIT_GUESS and OVERCOMMIT_NEVER.\n\nadmin_reserve_kbytes is initialized to min(3% free pages, 8MB)\n\nI arrived at 8MB by summing the RSS of sshd or login, bash, and top.\n\nPlease see first patch in this series for full background, motivation,\ntesting, and full changelog.\n\n[akpm@linux-foundation.org: coding-style fixes]\n[akpm@linux-foundation.org: make init_admin_reserve() static]\nSigned-off-by: Andrew Shewmaker \u003cagshew@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c9b1d0981fcce3d9976d7b7a56e4e0503bc610dd", "tree": "f7adfbfb1e34b3727e7d6b63a1ece93b358f87d8", "parents": [ "d8f10cb3d375c34ad668f32ca6e4661ad1fc23b2" ], "author": { "name": "Andrew Shewmaker", "email": "agshew@gmail.com", "time": "Mon Apr 29 15:08:10 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 29 15:54:36 2013 -0700" }, "message": "mm: limit growth of 3% hardcoded other user reserve\n\nAdd user_reserve_kbytes knob.\n\nLimit the growth of the memory reserved for other user processes to\nmin(3% current process size, user_reserve_pages). Only about 8MB is\nnecessary to enable recovery in the default mode, and only a few hundred\nMB are required even when overcommit is disabled.\n\nuser_reserve_pages defaults to min(3% free pages, 128MB)\n\nI arrived at 128MB by taking the max VSZ of sshd, login, bash, and top ...\nthen adding the RSS of each.\n\nThis only affects OVERCOMMIT_NEVER mode.\n\nBackground\n\n1. user reserve\n\n__vm_enough_memory reserves a hardcoded 3% of the current process size for\nother applications when overcommit is disabled. This was done so that a\nuser could recover if they launched a memory hogging process. Without the\nreserve, a user would easily run into a message such as:\n\nbash: fork: Cannot allocate memory\n\n2. admin reserve\n\nAdditionally, a hardcoded 3% of free memory is reserved for root in both\novercommit \u0027guess\u0027 and \u0027never\u0027 modes. This was intended to prevent a\nscenario where root-cant-log-in and perform recovery operations.\n\nNote that this reserve shrinks, and doesn\u0027t guarantee a useful reserve.\n\nMotivation\n\nThe two hardcoded memory reserves should be updated to account for current\nmemory sizes.\n\nAlso, the admin reserve would be more useful if it didn\u0027t shrink too much.\n\nWhen the current code was originally written, 1GB was considered\n\"enterprise\". Now the 3% reserve can grow to multiple GB on large memory\nsystems, and it only needs to be a few hundred MB at most to enable a user\nor admin to recover a system with an unwanted memory hogging process.\n\nI\u0027ve found that reducing these reserves is especially beneficial for a\nspecific type of application load:\n\n * single application system\n * one or few processes (e.g. one per core)\n * allocating all available memory\n * not initializing every page immediately\n * long running\n\nI\u0027ve run scientific clusters with this sort of load. A long running job\nsometimes failed many hours (weeks of CPU time) into a calculation. They\nweren\u0027t initializing all of their memory immediately, and they weren\u0027t\nusing calloc, so I put systems into overcommit \u0027never\u0027 mode. These\nclusters run diskless and have no swap.\n\nHowever, with the current reserves, a user wishing to allocate as much\nmemory as possible to one process may be prevented from using, for\nexample, almost 2GB out of 32GB.\n\nThe effect is less, but still significant when a user starts a job with\none process per core. I have repeatedly seen a set of processes\nrequesting the same amount of memory fail because one of them could not\nallocate the amount of memory a user would expect to be able to allocate.\nFor example, Message Passing Interfce (MPI) processes, one per core. And\nit is similar for other parallel programming frameworks.\n\nChanging this reserve code will make the overcommit never mode more useful\nby allowing applications to allocate nearly all of the available memory.\n\nAlso, the new admin_reserve_kbytes will be safer than the current behavior\nsince the hardcoded 3% of available memory reserve can shrink to something\nuseless in the case where applications have grabbed all available memory.\n\nRisks\n\n* \"bash: fork: Cannot allocate memory\"\n\n The downside of the first patch-- which creates a tunable user reserve\n that is only used in overcommit \u0027never\u0027 mode--is that an admin can set\n it so low that a user may not be able to kill their process, even if\n they already have a shell prompt.\n\n Of course, a user can get in the same predicament with the current 3%\n reserve--they just have to launch processes until 3% becomes negligible.\n\n* root-cant-log-in problem\n\n The second patch, adding the tunable rootuser_reserve_pages, allows\n the admin to shoot themselves in the foot by setting it too small. They\n can easily get the system into a state where root-can\u0027t-log-in.\n\n However, the new admin_reserve_kbytes will be safer than the current\n behavior since the hardcoded 3% of available memory reserve can shrink\n to something useless in the case where applications have grabbed all\n available memory.\n\nAlternatives\n\n * Memory cgroups provide a more flexible way to limit application memory.\n\n Not everyone wants to set up cgroups or deal with their overhead.\n\n * We could create a fourth overcommit mode which provides smaller reserves.\n\n The size of useful reserves may be drastically different depending\n on the whether the system is embedded or enterprise.\n\n * Force users to initialize all of their memory or use calloc.\n\n Some users don\u0027t want/expect the system to overcommit when they malloc.\n Overcommit \u0027never\u0027 mode is for this scenario, and it should work well.\n\nThe new user and admin reserve tunables are simple to use, with low\noverhead compared to cgroups. The patches preserve current behavior where\n3% of memory is less than 128MB, except that the admin reserve doesn\u0027t\nshrink to an unusable size under pressure. The code allows admins to tune\nfor embedded and enterprise usage.\n\nFAQ\n\n * How is the root-cant-login problem addressed?\n What happens if admin_reserve_pages is set to 0?\n\n Root is free to shoot themselves in the foot by setting\n admin_reserve_kbytes too low.\n\n On x86_64, the minimum useful reserve is:\n 8MB for overcommit \u0027guess\u0027\n 128MB for overcommit \u0027never\u0027\n\n admin_reserve_pages defaults to min(3% free memory, 8MB)\n\n So, anyone switching to \u0027never\u0027 mode needs to adjust\n admin_reserve_pages.\n\n * How do you calculate a minimum useful reserve?\n\n A user or the admin needs enough memory to login and perform\n recovery operations, which includes, at a minimum:\n\n sshd or login + bash (or some other shell) + top (or ps, kill, etc.)\n\n For overcommit \u0027guess\u0027, we can sum resident set sizes (RSS)\n because we only need enough memory to handle what the recovery\n programs will typically use. On x86_64 this is about 8MB.\n\n For overcommit \u0027never\u0027, we can take the max of their virtual sizes (VSZ)\n and add the sum of their RSS. We use VSZ instead of RSS because mode\n forces us to ensure we can fulfill all of the requested memory allocations--\n even if the programs only use a fraction of what they ask for.\n On x86_64 this is about 128MB.\n\n When swap is enabled, reserves are useful even when they are as\n small as 10MB, regardless of overcommit mode.\n\n When both swap and overcommit are disabled, then the admin should\n tune the reserves higher to be absolutley safe. Over 230MB each\n was safest in my testing.\n\n * What happens if user_reserve_pages is set to 0?\n\n Note, this only affects overcomitt \u0027never\u0027 mode.\n\n Then a user will be able to allocate all available memory minus\n admin_reserve_kbytes.\n\n However, they will easily see a message such as:\n\n \"bash: fork: Cannot allocate memory\"\n\n And they won\u0027t be able to recover/kill their application.\n The admin should be able to recover the system if\n admin_reserve_kbytes is set appropriately.\n\n * What\u0027s the difference between overcommit \u0027guess\u0027 and \u0027never\u0027?\n\n \"Guess\" allows an allocation if there are enough free + reclaimable\n pages. It has a hardcoded 3% of free pages reserved for root.\n\n \"Never\" allows an allocation if there is enough swap + a configurable\n percentage (default is 50) of physical RAM. It has a hardcoded 3% of\n free pages reserved for root, like \"Guess\" mode. It also has a\n hardcoded 3% of the current process size reserved for additional\n applications.\n\n * Why is overcommit \u0027guess\u0027 not suitable even when an app eventually\n writes to every page? It takes free pages, file pages, available\n swap pages, reclaimable slab pages into consideration. In other words,\n these are all pages available, then why isn\u0027t overcommit suitable?\n\n Because it only looks at the present state of the system. It\n does not take into account the memory that other applications have\n malloced, but haven\u0027t initialized yet. It overcommits the system.\n\nTest Summary\n\nThere was little change in behavior in the default overcommit \u0027guess\u0027\nmode with swap enabled before and after the patch. This was expected.\n\nSystems run most predictably (i.e. no oom kills) in overcommit \u0027never\u0027\nmode with swap enabled. This also allowed the most memory to be allocated\nto a user application.\n\nOvercommit \u0027guess\u0027 mode without swap is a bad idea. It is easy to\ncrash the system. None of the other tested combinations crashed.\nThis matches my experience on the Roadrunner supercomputer.\n\nWithout the tunable user reserve, a system in overcommit \u0027never\u0027 mode\nand without swap does not allow the admin to recover, although the\nadmin can.\n\nWith the new tunable reserves, a system in overcommit \u0027never\u0027 mode\nand without swap can be configured to:\n\n1. maximize user-allocatable memory, running close to the edge of\nrecoverability\n\n2. maximize recoverability, sacrificing allocatable memory to\nensure that a user cannot take down a system\n\nTest Description\n\nFedora 18 VM - 4 x86_64 cores, 5725MB RAM, 4GB Swap\n\nSystem is booted into multiuser console mode, with unnecessary services\nturned off. Caches were dropped before each test.\n\nHogs are user memtester processes that attempt to allocate all free memory\nas reported by /proc/meminfo\n\nIn overcommit \u0027never\u0027 mode, memory_ratio\u003d100\n\nTest Results\n\n3.9.0-rc1-mm1\n\nOvercommit | Swap | Hogs | MB Got/Wanted | OOMs | User Recovery | Admin Recovery\n---------- ---- ---- ------------- ---- ------------- --------------\nguess yes 1 5432/5432 no yes yes\nguess yes 4 5444/5444 1 yes yes\nguess no 1 5302/5449 no yes yes\nguess no 4 - crash no no\n\nnever yes 1 5460/5460 1 yes yes\nnever yes 4 5460/5460 1 yes yes\nnever no 1 5218/5432 no no yes\nnever no 4 5203/5448 no no yes\n\n3.9.0-rc1-mm1-tunablereserves\n\nUser and Admin Recovery show their respective reserves, if applicable.\n\nOvercommit | Swap | Hogs | MB Got/Wanted | OOMs | User Recovery | Admin Recovery\n---------- ---- ---- ------------- ---- ------------- --------------\nguess yes 1 5419/5419 no - yes 8MB yes\nguess yes 4 5436/5436 1 - yes 8MB yes\nguess no 1 5440/5440 * - yes 8MB yes\nguess no 4 - crash - no 8MB no\n\n* process would successfully mlock, then the oom killer would pick it\n\nnever yes 1 5446/5446 no 10MB yes 20MB yes\nnever yes 4 5456/5456 no 10MB yes 20MB yes\nnever no 1 5387/5429 no 128MB no 8MB barely\nnever no 1 5323/5428 no 226MB barely 8MB barely\nnever no 1 5323/5428 no 226MB barely 8MB barely\n\nnever no 1 5359/5448 no 10MB no 10MB barely\n\nnever no 1 5323/5428 no 0MB no 10MB barely\nnever no 1 5332/5428 no 0MB no 50MB yes\nnever no 1 5293/5429 no 0MB no 90MB yes\n\nnever no 1 5001/5427 no 230MB yes 338MB yes\nnever no 4* 4998/5424 no 230MB yes 338MB yes\n\n* more memtesters were launched, able to allocate approximately another 100MB\n\nFuture Work\n\n - Test larger memory systems.\n\n - Test an embedded image.\n\n - Test other architectures.\n\n - Time malloc microbenchmarks.\n\n - Would it be useful to be able to set overcommit policy for\n each memory cgroup?\n\n - Some lines are slightly above 80 chars.\n Perhaps define a macro to convert between pages and kb?\n Other places in the kernel do this.\n\n[akpm@linux-foundation.org: coding-style fixes]\n[akpm@linux-foundation.org: make init_user_reserve() static]\nSigned-off-by: Andrew Shewmaker \u003cagshew@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "146732ce104ddfed3d4d82722c0b336074016b92", "tree": "07185bc6ff143cc669cc9e0a197a3adf0c13004a", "parents": [ "6d2488f64a240191f0733c1f32d73607916b01b7" ], "author": { "name": "Josh Triplett", "email": "josh@joshtriplett.org", "time": "Mon Apr 29 15:07:22 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 29 15:54:33 2013 -0700" }, "message": "fs: don\u0027t compile in drop_caches.c when CONFIG_SYSCTL\u003dn\n\ndrop_caches.c provides code only invokable via sysctl, so don\u0027t compile it\nin when CONFIG_SYSCTL\u003dn.\n\nSigned-off-by: Josh Triplett \u003cjosh@joshtriplett.org\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e23b62256a361611cbd45cd1456638f1a5106b5c", "tree": "472968c961432a1d9d0c3634ca20433f1d9cd29b", "parents": [ "aebb2afd5420c860b7fbc3882a323ef1247fbf16", "8ccfe6675fa974bd06d64f74d0fdee6a5267d2aa" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 02 07:58:56 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 02 07:58:56 2013 -0800" }, "message": "Merge tag \u0027arc-v3.9-rc1-late\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc\n\nPull new ARC architecture from Vineet Gupta:\n \"Initial ARC Linux port with some fixes on top for 3.9-rc1:\n\n I would like to introduce the Linux port to ARC Processors (from\n Synopsys) for 3.9-rc1. The patch-set has been discussed on the public\n lists since Nov and has received a fair bit of review, specially from\n Arnd, tglx, Al and other subsystem maintainers for DeviceTree, kgdb...\n\n The arch bits are in arch/arc, some asm-generic changes (acked by\n Arnd), a minor change to PARISC (acked by Helge).\n\n The series is a touch bigger for a new port for 2 main reasons:\n\n 1. It enables a basic kernel in first sub-series and adds\n ptrace/kgdb/.. later\n\n 2. Some of the fallout of review (DeviceTree support, multi-platform-\n image support) were added on top of orig series, primarily to\n record the revision history.\n\n This updated pull request additionally contains\n\n - fixes due to our GNU tools catching up with the new syscall/ptrace\n ABI\n\n - some (minor) cross-arch Kconfig updates.\"\n\n* tag \u0027arc-v3.9-rc1-late\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc: (82 commits)\n ARC: split elf.h into uapi and export it for userspace\n ARC: Fixup the current ABI version\n ARC: gdbserver using regset interface possibly broken\n ARC: Kconfig cleanup tracking cross-arch Kconfig pruning in merge window\n ARC: make a copy of flat DT\n ARC: [plat-arcfpga] DT arc-uart bindings change: \"baud\" \u003d\u003e \"current-speed\"\n ARC: Ensure CONFIG_VIRT_TO_BUS is not enabled\n ARC: Fix pt_orig_r8 access\n ARC: [3.9] Fallout of hlist iterator update\n ARC: 64bit RTSC timestamp hardware issue\n ARC: Don\u0027t fiddle with non-existent caches\n ARC: Add self to MAINTAINERS\n ARC: Provide a default serial.h for uart drivers needing BASE_BAUD\n ARC: [plat-arcfpga] defconfig for fully loaded ARC Linux\n ARC: [Review] Multi-platform image #8: platform registers SMP callbacks\n ARC: [Review] Multi-platform image #7: SMP common code to use callbacks\n ARC: [Review] Multi-platform image #6: cpu-to-dma-addr optional\n ARC: [Review] Multi-platform image #5: NR_IRQS defined by ARC core\n ARC: [Review] Multi-platform image #4: Isolate platform headers\n ARC: [Review] Multi-platform image #3: switch to board callback\n ...\n" }, { "commit": "e579d2c259be42b6f29458327e5153b22414b031", "tree": "ac3babd0f075ddfeb5e6fa083651852b34a0a79f", "parents": [ "5d1fadc1472396d602f0eeb10d37519e2a14e8bc" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Feb 27 17:03:15 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 27 19:10:11 2013 -0800" }, "message": "coredump: remove redundant defines for dumpable states\n\nThe existing SUID_DUMP_* defines duplicate the newer SUID_DUMPABLE_*\ndefines introduced in 54b501992dd2 (\"coredump: warn about unsafe\nsuid_dumpable / core_pattern combo\"). Remove the new ones, and use the\nprior values instead.\n\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReported-by: Chen Gang \u003cgang.chen@asianux.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Alan Cox \u003calan@linux.intel.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Doug Ledford \u003cdledford@redhat.com\u003e\nCc: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nCc: James Morris \u003cjames.l.morris@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9043a2650cd21f96f831a97f516c2c302e21fb70", "tree": "926720afb0acc7bad8cfcae537dc58de552f9249", "parents": [ "ab7826595e9ec51a51f622c5fc91e2f59440481a", "d9d8d7ed498ec65bea72dd24be7b9cd35af0c200" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 25 15:41:43 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 25 15:41:43 2013 -0800" }, "message": "Merge tag \u0027modules-next-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux\n\nPull module update from Rusty Russell:\n \"The sweeping change is to make add_taint() explicitly indicate whether\n to disable lockdep, but it\u0027s a mechanical change.\"\n\n* tag \u0027modules-next-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:\n MODSIGN: Add option to not sign modules during modules_install\n MODSIGN: Add -s \u003csignature\u003e option to sign-file\n MODSIGN: Specify the hash algorithm on sign-file command line\n MODSIGN: Simplify Makefile with a Kconfig helper\n module: clean up load_module a little more.\n modpost: Ignore ARC specific non-alloc sections\n module: constify within_module_*\n taint: add explicit flag to show whether lock dep is still OK.\n module: printk message when module signature fail taints kernel.\n" }, { "commit": "75f7ad8e043d9383337d917584297f7737154bbf", "tree": "101f5f2373b615bf83838cd13615fd123dfd765c", "parents": [ "08b52706d505658eac0962d215ff697f898bbc13" ], "author": { "name": "Paul Szabo", "email": "psz@maths.usyd.edu.au", "time": "Fri Feb 22 16:34:42 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 23 17:50:17 2013 -0800" }, "message": "page-writeback.c: subtract min_free_kbytes from dirtyable memory\n\nWhen calculating amount of dirtyable memory, min_free_kbytes should be\nsubtracted because it is not intended for dirty pages.\n\nAddresses http://bugs.debian.org/695182\n\n[akpm@linux-foundation.org: fix up min_free_kbytes extern declarations]\n[akpm@linux-foundation.org: fix min() warning]\nSigned-off-by: Paul Szabo \u003cpsz@maths.usyd.edu.au\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "27ea6dfdc23e5e96e324d6cb3488528c14e4a7f7", "tree": "f3a77ed29e1cecd1f401b86384b322f1b005ef0b", "parents": [ "81ec44a6c69342fec1b1140c60a604027e429f69", "cc883afcc02e1c9c8ab64d20f0288355b857f966" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 21 17:55:48 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 21 17:55:48 2013 -0800" }, "message": "Merge tag \u0027please-pull-misc-3.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux\n\nPull misc ia64 bits from Tony Luck.\n\n* tag \u0027please-pull-misc-3.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux:\n MAINTAINERS: update SGI \u0026 ia64 Altix stuff\n sysctl: Enable IA64 \"ignore-unaligned-usertrap\" to be used cross-arch\n" }, { "commit": "bf14e3b979a01cd7298d631736f965fe83c6e2bc", "tree": "8440a0641556332763615edc881f81d1da6f2d37", "parents": [ "4d86dfbbda09b3c67bcaeb370f22a2cc7f39205b" ], "author": { "name": "Vineet Gupta", "email": "vgupta@synopsys.com", "time": "Fri Jan 18 15:12:24 2013 +0530" }, "committer": { "name": "Vineet Gupta", "email": "vgupta@synopsys.com", "time": "Fri Feb 15 23:16:05 2013 +0530" }, "message": "sysctl: Enable PARISC \"unaligned-trap\" to be used cross-arch\n\nPARISC defines /proc/sys/kernel/unaligned-trap to runtime toggle\nunaligned access emulation.\n\nThe exact mechanics of enablig/disabling are still arch specific, we can\nmake the sysctl usable by other arches.\n\nSigned-off-by: Vineet Gupta \u003cvgupta@synopsys.com\u003e\nAcked-by: Helge Deller \u003cdeller@gmx.de\u003e\nCc: \"James E.J. Bottomley\" \u003cjejb@parisc-linux.org\u003e\nCc: Helge Deller \u003cdeller@gmx.de\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\n" }, { "commit": "ce0dbbbb30aee6a835511d5be446462388ba9eee", "tree": "c3d4d6f64a0fe7eec9283f4413d111a8d37d7053", "parents": [ "cf4aebc292fac7f34f8345664320e9d4a42ca76c" ], "author": { "name": "Clark Williams", "email": "williams@redhat.com", "time": "Thu Feb 07 09:47:04 2013 -0600" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Thu Feb 07 20:51:07 2013 +0100" }, "message": "sched/rt: Add a tuning knob to allow changing SCHED_RR timeslice\n\nAdd a /proc/sys/kernel scheduler knob named\nsched_rr_timeslice_ms that allows global changing of the\nSCHED_RR timeslice value. User visable value is in milliseconds\nbut is stored as jiffies. Setting to 0 (zero) resets to the\ndefault (currently 100ms).\n\nSigned-off-by: Clark Williams \u003cwilliams@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nLink: http://lkml.kernel.org/r/20130207094704.13751796@riff.lan\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "cf4aebc292fac7f34f8345664320e9d4a42ca76c", "tree": "6eceb9bb2d8382c4499366a8fee060688aad6107", "parents": [ "b2c77a57e4a0a7877e357dead7ee8acc19944f3e" ], "author": { "name": "Clark Williams", "email": "williams@redhat.com", "time": "Thu Feb 07 09:46:59 2013 -0600" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Thu Feb 07 20:50:54 2013 +0100" }, "message": "sched: Move sched.h sysctl bits into separate header\n\nMove the sysctl-related bits from include/linux/sched.h into\na new file: include/linux/sched/sysctl.h. Then update source\nfiles requiring access to those bits by including the new\nheader file.\n\nSigned-off-by: Clark Williams \u003cwilliams@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nLink: http://lkml.kernel.org/r/20130207094659.06dced96@riff.lan\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "373d4d099761cb1f637bed488ab3871945882273", "tree": "954bef7bc724aee105dd246d5f2b1ea04ed38b20", "parents": [ "64748a2c9062da0c32b59c1b368a86fc4613b1e1" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Mon Jan 21 17:17:39 2013 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Mon Jan 21 17:17:57 2013 +1030" }, "message": "taint: add explicit flag to show whether lock dep is still OK.\n\nFix up all callers as they were before, with make one change: an\nunsigned module taints the kernel, but doesn\u0027t turn off lockdep.\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\n" }, { "commit": "b6fca72536fb94098acb23c00b4e65d3bc4bb252", "tree": "5fec671de29e4fee7a0be6aca6e93d0acbe09a0d", "parents": [ "d1c3ed669a2d452cacfb48c2d171a1f364dae2ed" ], "author": { "name": "Vineet Gupta", "email": "Vineet.Gupta1@synopsys.com", "time": "Wed Jan 09 20:06:28 2013 +0530" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jan 09 10:48:34 2013 -0800" }, "message": "sysctl: Enable IA64 \"ignore-unaligned-usertrap\" to be used cross-arch\n\nIA64 defines /proc/sys/kernel/ignore-unaligned-usertrap to control\nverbose warnings on unaligned access emulation.\n\nAlthough the exact mechanics of what to do with sysctl (ignore/shout)\nare arch specific, this change enables the sysctl to be usable cross-arch.\n\nSigned-off-by: Vineet Gupta \u003cvgupta@synopsys.com\u003e\nCc: Fenghua Yu \u003cfenghua.yu@intel.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "3d59eebc5e137bd89c6351e4c70e90ba1d0dc234", "tree": "b4ddfd0b057454a7437a3b4e3074a3b8b4b03817", "parents": [ "11520e5e7c1855fc3bf202bb3be35a39d9efa034", "4fc3f1d66b1ef0d7b8dc11f4ff1cc510f78b37d6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Dec 16 14:33:25 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Dec 16 15:18:08 2012 -0800" }, "message": "Merge tag \u0027balancenuma-v11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma\n\nPull Automatic NUMA Balancing bare-bones from Mel Gorman:\n \"There are three implementations for NUMA balancing, this tree\n (balancenuma), numacore which has been developed in tip/master and\n autonuma which is in aa.git.\n\n In almost all respects balancenuma is the dumbest of the three because\n its main impact is on the VM side with no attempt to be smart about\n scheduling. In the interest of getting the ball rolling, it would be\n desirable to see this much merged for 3.8 with the view to building\n scheduler smarts on top and adapting the VM where required for 3.9.\n\n The most recent set of comparisons available from different people are\n\n mel: https://lkml.org/lkml/2012/12/9/108\n mingo: https://lkml.org/lkml/2012/12/7/331\n tglx: https://lkml.org/lkml/2012/12/10/437\n srikar: https://lkml.org/lkml/2012/12/10/397\n\n The results are a mixed bag. In my own tests, balancenuma does\n reasonably well. It\u0027s dumb as rocks and does not regress against\n mainline. On the other hand, Ingo\u0027s tests shows that balancenuma is\n incapable of converging for this workloads driven by perf which is bad\n but is potentially explained by the lack of scheduler smarts. Thomas\u0027\n results show balancenuma improves on mainline but falls far short of\n numacore or autonuma. Srikar\u0027s results indicate we all suffer on a\n large machine with imbalanced node sizes.\n\n My own testing showed that recent numacore results have improved\n dramatically, particularly in the last week but not universally.\n We\u0027ve butted heads heavily on system CPU usage and high levels of\n migration even when it shows that overall performance is better.\n There are also cases where it regresses. Of interest is that for\n specjbb in some configurations it will regress for lower numbers of\n warehouses and show gains for higher numbers which is not reported by\n the tool by default and sometimes missed in treports. Recently I\n reported for numacore that the JVM was crashing with\n NullPointerExceptions but currently it\u0027s unclear what the source of\n this problem is. Initially I thought it was in how numacore batch\n handles PTEs but I\u0027m no longer think this is the case. It\u0027s possible\n numacore is just able to trigger it due to higher rates of migration.\n\n These reports were quite late in the cycle so I/we would like to start\n with this tree as it contains much of the code we can agree on and has\n not changed significantly over the last 2-3 weeks.\"\n\n* tag \u0027balancenuma-v11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma: (50 commits)\n mm/rmap, migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable\n mm/rmap: Convert the struct anon_vma::mutex to an rwsem\n mm: migrate: Account a transhuge page properly when rate limiting\n mm: numa: Account for failed allocations and isolations as migration failures\n mm: numa: Add THP migration for the NUMA working set scanning fault case build fix\n mm: numa: Add THP migration for the NUMA working set scanning fault case.\n mm: sched: numa: Delay PTE scanning until a task is scheduled on a new node\n mm: sched: numa: Control enabling and disabling of NUMA balancing if !SCHED_DEBUG\n mm: sched: numa: Control enabling and disabling of NUMA balancing\n mm: sched: Adapt the scanning rate if a NUMA hinting fault does not migrate\n mm: numa: Use a two-stage filter to restrict pages being migrated for unlikely task\u003c-\u003enode relationships\n mm: numa: migrate: Set last_nid on newly allocated page\n mm: numa: split_huge_page: Transfer last_nid on tail page\n mm: numa: Introduce last_nid to the page frame\n sched: numa: Slowly increase the scanning period as NUMA faults are handled\n mm: numa: Rate limit setting of pte_numa if node is saturated\n mm: numa: Rate limit the amount of memory that is migrated between nodes\n mm: numa: Structures for Migrate On Fault per NUMA migration rate limiting\n mm: numa: Migrate pages handled during a pmd_numa hinting fault\n mm: numa: Migrate on reference policy\n ...\n" }, { "commit": "b8593bfda1652755136333cdd362de125b283a9c", "tree": "c0395d9cf775fd9225e81b055fc8f5540a14333a", "parents": [ "e42c8ff2999de1239a57d434bfbd8e9f2a56e814" ], "author": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Wed Nov 21 01:18:23 2012 +0000" }, "committer": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Tue Dec 11 14:42:55 2012 +0000" }, "message": "mm: sched: Adapt the scanning rate if a NUMA hinting fault does not migrate\n\nThe PTE scanning rate and fault rates are two of the biggest sources of\nsystem CPU overhead with automatic NUMA placement. Ideally a proper policy\nwould detect if a workload was properly placed, schedule and adjust the\nPTE scanning rate accordingly. We do not track the necessary information\nto do that but we at least know if we migrated or not.\n\nThis patch scans slower if a page was not migrated as the result of a\nNUMA hinting fault up to sysctl_numa_balancing_scan_period_max which is\nnow higher than the previous default. Once every minute it will reset\nthe scanner in case of phase changes.\n\nThis is hilariously crude and the numbers are arbitrary. Workloads will\nconverge quite slowly in comparison to what a proper policy should be able\nto do. On the plus side, we will chew up less CPU for workloads that have\nno need for automatic balancing.\n\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\n" }, { "commit": "4b96a29ba891dd59734cb7be80a900fe93aa2d9f", "tree": "5162223c4ceae37f6ccf0ef1b84993c2556e60cf", "parents": [ "9f40604cdab935e80db57b309c48659de349d4e6" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Thu Oct 25 14:16:47 2012 +0200" }, "committer": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Tue Dec 11 14:42:47 2012 +0000" }, "message": "mm: sched: numa: Implement slow start for working set sampling\n\nAdd a 1 second delay before starting to scan the working set of\na task and starting to balance it amongst nodes.\n\n[ note that before the constant per task WSS sampling rate patch\n the initial scan would happen much later still, in effect that\n patch caused this regression. ]\n\nThe theory is that short-run tasks benefit very little from NUMA\nplacement: they come and go, and they better stick to the node\nthey were started on. As tasks mature and rebalance to other CPUs\nand nodes, so does their NUMA placement have to change and so\ndoes it start to matter more and more.\n\nIn practice this change fixes an observable kbuild regression:\n\n # [ a perf stat --null --repeat 10 test of ten bzImage builds to /dev/shm ]\n\n !NUMA:\n 45.291088843 seconds time elapsed ( +- 0.40% )\n 45.154231752 seconds time elapsed ( +- 0.36% )\n\n +NUMA, no slow start:\n 46.172308123 seconds time elapsed ( +- 0.30% )\n 46.343168745 seconds time elapsed ( +- 0.25% )\n\n +NUMA, 1 sec slow start:\n 45.224189155 seconds time elapsed ( +- 0.25% )\n 45.160866532 seconds time elapsed ( +- 0.17% )\n\nand it also fixes an observable perf bench (hackbench) regression:\n\n # perf stat --null --repeat 10 perf bench sched messaging\n\n -NUMA:\n\n -NUMA: 0.246225691 seconds time elapsed ( +- 1.31% )\n +NUMA no slow start: 0.252620063 seconds time elapsed ( +- 1.13% )\n\n +NUMA 1sec delay: 0.248076230 seconds time elapsed ( +- 1.35% )\n\nThe implementation is simple and straightforward, most of the patch\ndeals with adding the /proc/sys/kernel/numa_balancing_scan_delay_ms tunable\nknob.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\n[ Wrote the changelog, ran measurements, tuned the default. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\n" }, { "commit": "6e5fb223e89dbe5cb5c563f8d4a4a0a7d62455a8", "tree": "0d5c93240702a51b1d6f22fefd979235a19692fd", "parents": [ "cbee9f88ec1b8dd6b58f25f54e4f52c82ed77690" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Thu Oct 25 14:16:45 2012 +0200" }, "committer": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Tue Dec 11 14:42:46 2012 +0000" }, "message": "mm: sched: numa: Implement constant, per task Working Set Sampling (WSS) rate\n\nPreviously, to probe the working set of a task, we\u0027d use\na very simple and crude method: mark all of its address\nspace PROT_NONE.\n\nThat method has various (obvious) disadvantages:\n\n - it samples the working set at dissimilar rates,\n giving some tasks a sampling quality advantage\n over others.\n\n - creates performance problems for tasks with very\n large working sets\n\n - over-samples processes with large address spaces but\n which only very rarely execute\n\nImprove that method by keeping a rotating offset into the\naddress space that marks the current position of the scan,\nand advance it by a constant rate (in a CPU cycles execution\nproportional manner). If the offset reaches the last mapped\naddress of the mm then it then it starts over at the first\naddress.\n\nThe per-task nature of the working set sampling functionality in this tree\nallows such constant rate, per task, execution-weight proportional sampling\nof the working set, with an adaptive sampling interval/frequency that\ngoes from once per 100ms up to just once per 8 seconds. The current\nsampling volume is 256 MB per interval.\n\nAs tasks mature and converge their working set, so does the\nsampling rate slow down to just a trickle, 256 MB per 8\nseconds of CPU time executed.\n\nThis, beyond being adaptive, also rate-limits rarely\nexecuting systems and does not over-sample on overloaded\nsystems.\n\n[ In AutoNUMA speak, this patch deals with the effective sampling\n rate of the \u0027hinting page fault\u0027. AutoNUMA\u0027s scanning is\n currently rate-limited, but it is also fundamentally\n single-threaded, executing in the knuma_scand kernel thread,\n so the limit in AutoNUMA is global and does not scale up with\n the number of CPUs, nor does it scan tasks in an execution\n proportional manner.\n\n So the idea of rate-limiting the scanning was first implemented\n in the AutoNUMA tree via a global rate limit. This patch goes\n beyond that by implementing an execution rate proportional\n working set sampling rate that is not implemented via a single\n global scanning daemon. ]\n\n[ Dan Carpenter pointed out a possible NULL pointer dereference in the\n first version of this patch. ]\n\nBased-on-idea-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nBug-Found-By: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\n[ Wrote changelog and fixed bug. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\n" }, { "commit": "cbee9f88ec1b8dd6b58f25f54e4f52c82ed77690", "tree": "d4cfbcfa3e89742216cd792d4aa914356406b532", "parents": [ "a720094ded8cbb303111035be91858011d2eac71" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Thu Oct 25 14:16:43 2012 +0200" }, "committer": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Tue Dec 11 14:42:45 2012 +0000" }, "message": "mm: numa: Add fault driven placement and migration\n\nNOTE: This patch is based on \"sched, numa, mm: Add fault driven\n\tplacement and migration policy\" but as it throws away all the policy\n\tto just leave a basic foundation I had to drop the signed-offs-by.\n\nThis patch creates a bare-bones method for setting PTEs pte_numa in the\ncontext of the scheduler that when faulted later will be faulted onto the\nnode the CPU is running on. In itself this does nothing useful but any\nplacement policy will fundamentally depend on receiving hints on placement\nfrom fault context and doing something intelligent about it.\n\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\n" }, { "commit": "3b572b506c76a7a38c8debe45b50d09295620810", "tree": "f1a6e47c3eadc04f44f3ba086712ad853cea992a", "parents": [ "1c9a9f59149e247e264db1fbf9f8ea3d5066eb64" ], "author": { "name": "Bill Pemberton", "email": "wfp5p@virginia.edu", "time": "Mon Nov 19 13:19:29 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Wed Nov 28 10:33:03 2012 -0800" }, "message": "sysctl: remove CONFIG_HOTPLUG ifdefs\n\nRemove conditional code based on CONFIG_HOTPLUG being false. It\u0027s\nalways on now in preparation of it going away as an option.\n\nSigned-off-by: Bill Pemberton \u003cwfp5p@virginia.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "7ac57a89de958fbb5271dc504d0c25e34dbeec32", "tree": "4243b7e9f702e9b48820b62d8572f9aeb6f1cf1c", "parents": [ "9b2a60c484715e2d2f07d8192fd9f18435cbc77c" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Mon Oct 08 16:28:16 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 09 16:22:14 2012 +0900" }, "message": "Kconfig: clean up the \"#if defined(arch)\" list for exception-trace sysctl entry\n\nIntroduce SYSCTL_EXCEPTION_TRACE config option and selec it in the\narchitectures requiring support for the \"exception-trace\" debug_table\nentry in kernel/sysctl.c.\n\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Chris Metcalf \u003ccmetcalf@tilera.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "046d662f481830e652ac34cd112249adde16452a", "tree": "ad13e968b4b02f9e2c35ce45f358477474df80d7", "parents": [ "db9aeca97a58563e1ab927d157c9b5048f233e73" ], "author": { "name": "Alex Kelly", "email": "alex.page.kelly@gmail.com", "time": "Thu Oct 04 17:15:23 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Oct 06 03:05:15 2012 +0900" }, "message": "coredump: make core dump functionality optional\n\nAdds an expert Kconfig option, CONFIG_COREDUMP, which allows disabling of\ncore dump. This saves approximately 2.6k in the compiled kernel, and\ncomplements CONFIG_ELF_CORE, which now depends on it.\n\nCONFIG_COREDUMP also disables coredump-related sysctls, except for\nsuid_dumpable and related functions, which are necessary for ptrace.\n\n[akpm@linux-foundation.org: fix binfmt_aout.c build]\nSigned-off-by: Alex Kelly \u003calex.page.kelly@gmail.com\u003e\nReviewed-by: Josh Triplett \u003cjosh@joshtriplett.org\u003e\nAcked-by: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "81f56e5375e84689b891e0e6c5a02ec12a1f18d9", "tree": "a1e128a71ff24fc705428df86a858076cfe4bc13", "parents": [ "6c09931b3f987898f5c581d267ef269f5e2e9575", "27aa55c5e5123fa8b8ad0156559d34d7edff58ca" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 01 11:51:57 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 01 11:51:57 2012 -0700" }, "message": "Merge tag \u0027arm64-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64\n\nPull arm64 support from Catalin Marinas:\n \"Linux support for the 64-bit ARM architecture (AArch64)\n\n Features currently supported:\n - 39-bit address space for user and kernel (each)\n - 4KB and 64KB page configurations\n - Compat (32-bit) user applications (ARMv7, EABI only)\n - Flattened Device Tree (mandated for all AArch64 platforms)\n - ARM generic timers\"\n\n* tag \u0027arm64-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64: (35 commits)\n arm64: ptrace: remove obsolete ptrace request numbers from user headers\n arm64: Do not set the SMP/nAMP processor bit\n arm64: MAINTAINERS update\n arm64: Build infrastructure\n arm64: Miscellaneous header files\n arm64: Generic timers support\n arm64: Loadable modules\n arm64: Miscellaneous library functions\n arm64: Performance counters support\n arm64: Add support for /proc/sys/debug/exception-trace\n arm64: Debugging support\n arm64: Floating point and SIMD\n arm64: 32-bit (compat) applications support\n arm64: User access library functions\n arm64: Signal handling support\n arm64: VDSO support\n arm64: System calls handling\n arm64: ELF definitions\n arm64: SMP support\n arm64: DMA mapping API\n ...\n" }, { "commit": "5c4233697c3f5cb14eb7a969332e2d60f357f952", "tree": "9708b02810a1007b6ba9c54eddd2f62eb9924c14", "parents": [ "478fcb2cdb2351dcfc3fb23f42d76f4436ee4149" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Tue Aug 14 17:08:45 2012 +0100" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Mon Sep 17 13:42:16 2012 +0100" }, "message": "arm64: Add support for /proc/sys/debug/exception-trace\n\nThis patch allows setting of the show_unhandled_signals variable via\n/proc/sys/debug/exception-trace. The default value is currently 1\nshowing unhandled user faults (undefined instructions, data aborts) and\ninvalid signal stack frames.\n\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nAcked-by: Tony Lindgren \u003ctony@atomide.com\u003e\nAcked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nAcked-by: Olof Johansson \u003colof@lixom.net\u003e\nAcked-by: Santosh Shilimkar \u003csantosh.shilimkar@ti.com\u003e\n" }, { "commit": "d00535db42805e9ae5eadf1b4a86e01e85674b0c", "tree": "808a6058e6b27c9ef676ed72590d94c077983bb8", "parents": [ "201c373e8e4823700d3160d5c28e1ab18fd1193e" ], "author": { "name": "Namhyung Kim", "email": "namhyung.kim@lge.com", "time": "Thu Aug 16 11:15:30 2012 +0900" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Sep 04 14:31:34 2012 +0200" }, "message": "sched: Add time unit suffix to sched sysctl knobs\n\nUnlike others, sched_migration_cost, sched_time_avg and\nsched_shares_window doesn\u0027t have time unit as suffix. Add them.\n\nSigned-off-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLink: http://lkml.kernel.org/r/1345083330-19486-1-git-send-email-namhyung@kernel.org\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "a0e881b7c189fa2bd76c024dbff91e79511c971d", "tree": "0c801918565b08921d21aceee5b326f64d998f5f", "parents": [ "eff0d13f3823f35d70228cd151d2a2c89288ff32", "dbc6e0222d79e78925fe20733844a796a4b72cf9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 01 10:26:23 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 01 10:26:23 2012 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull second vfs pile from Al Viro:\n \"The stuff in there: fsfreeze deadlock fixes by Jan (essentially, the\n deadlock reproduced by xfstests 068), symlink and hardlink restriction\n patches, plus assorted cleanups and fixes.\n\n Note that another fsfreeze deadlock (emergency thaw one) is *not*\n dealt with - the series by Fernando conflicts a lot with Jan\u0027s, breaks\n userland ABI (FIFREEZE semantics gets changed) and trades the deadlock\n for massive vfsmount leak; this is going to be handled next cycle.\n There probably will be another pull request, but that stuff won\u0027t be\n in it.\"\n\nFix up trivial conflicts due to unrelated changes next to each other in\ndrivers/{staging/gdm72xx/usb_boot.c, usb/gadget/storage_common.c}\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (54 commits)\n delousing target_core_file a bit\n Documentation: Correct s_umount state for freeze_fs/unfreeze_fs\n fs: Remove old freezing mechanism\n ext2: Implement freezing\n btrfs: Convert to new freezing mechanism\n nilfs2: Convert to new freezing mechanism\n ntfs: Convert to new freezing mechanism\n fuse: Convert to new freezing mechanism\n gfs2: Convert to new freezing mechanism\n ocfs2: Convert to new freezing mechanism\n xfs: Convert to new freezing code\n ext4: Convert to new freezing mechanism\n fs: Protect write paths by sb_start_write - sb_end_write\n fs: Skip atime update on frozen filesystem\n fs: Add freezing handling to mnt_want_write() / mnt_drop_write()\n fs: Improve filesystem freezing handling\n switch the protection of percpu_counter list to spinlock\n nfsd: Push mnt_want_write() outside of i_mutex\n btrfs: Push mnt_want_write() outside of i_mutex\n fat: Push mnt_want_write() outside of i_mutex\n ...\n" }, { "commit": "3965c9ae47d64aadf6f13b6fcd37767b83c0689a", "tree": "9896fbd841edbaa8e374b593ad0f7b96191cda5d", "parents": [ "deaf386ee58d5336bbef8959bf304573afb67c20" ], "author": { "name": "Wanpeng Li", "email": "liwp@linux.vnet.ibm.com", "time": "Tue Jul 31 16:41:52 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 31 18:42:40 2012 -0700" }, "message": "mm: prepare for removal of obsolete /proc/sys/vm/nr_pdflush_threads\n\nSince per-BDI flusher threads were introduced in 2.6, the pdflush\nmechanism is not used any more. But the old interface exported through\n/proc/sys/vm/nr_pdflush_threads still exists and is obviously useless.\n\nFor back-compatibility, printk warning information and return 2 to notify\nthe users that the interface is removed.\n\nSigned-off-by: Wanpeng Li \u003cliwp@linux.vnet.ibm.com\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fd4b616b0fbb77e3f349e7d60914f2b7c7e39f9c", "tree": "a0cc595c7680bd08467fb803738d534cfa8f150e", "parents": [ "c1d7e01d7877a397655277a920aeaa3830ed9461" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Jul 30 14:42:48 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 30 17:25:21 2012 -0700" }, "message": "sysctl: suppress kmemleak messages\n\nregister_sysctl_table() is a strange function, as it makes internal\nallocations (a header) to register a sysctl_table. This header is a\nhandle to the table that is created, and can be used to unregister the\ntable. But if the table is permanent and never unregistered, the header\nacts the same as a static variable.\n\nUnfortunately, this allocation of memory that is never expected to be\nfreed fools kmemleak in thinking that we have leaked memory. For those\nsysctl tables that are never unregistered, and have no pointer referencing\nthem, kmemleak will think that these are memory leaks:\n\nunreferenced object 0xffff880079fb9d40 (size 192):\n comm \"swapper/0\", pid 0, jiffies 4294667316 (age 12614.152s)\n hex dump (first 32 bytes):\n 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n backtrace:\n [\u003cffffffff8146b590\u003e] kmemleak_alloc+0x73/0x98\n [\u003cffffffff8110a935\u003e] kmemleak_alloc_recursive.constprop.42+0x16/0x18\n [\u003cffffffff8110b852\u003e] __kmalloc+0x107/0x153\n [\u003cffffffff8116fa72\u003e] kzalloc.constprop.8+0xe/0x10\n [\u003cffffffff811703c9\u003e] __register_sysctl_paths+0xe1/0x160\n [\u003cffffffff81170463\u003e] register_sysctl_paths+0x1b/0x1d\n [\u003cffffffff8117047d\u003e] register_sysctl_table+0x18/0x1a\n [\u003cffffffff81afb0a1\u003e] sysctl_init+0x10/0x14\n [\u003cffffffff81b05a6f\u003e] proc_sys_init+0x2f/0x31\n [\u003cffffffff81b0584c\u003e] proc_root_init+0xa5/0xa7\n [\u003cffffffff81ae5b7e\u003e] start_kernel+0x3d0/0x40a\n [\u003cffffffff81ae52a7\u003e] x86_64_start_reservations+0xae/0xb2\n [\u003cffffffff81ae53ad\u003e] x86_64_start_kernel+0x102/0x111\n [\u003cffffffffffffffff\u003e] 0xffffffffffffffff\n\nThe sysctl_base_table used by sysctl itself is one such instance that\nregisters the table to never be unregistered.\n\nUse kmemleak_not_leak() to suppress the kmemleak false positive.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nAcked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "54b501992dd2a839e94e76aa392c392b55080ce8", "tree": "6d174fb95516f01e385d31d5ef87a248d8740fd2", "parents": [ "9520628e8ceb69fa9a4aee6b57f22675d9e1b709" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Mon Jul 30 14:39:18 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 30 17:25:11 2012 -0700" }, "message": "coredump: warn about unsafe suid_dumpable / core_pattern combo\n\nWhen suid_dumpable\u003d2, detect unsafe core_pattern settings and warn when\nthey are seen.\n\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSuggested-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Alan Cox \u003calan@linux.intel.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Doug Ledford \u003cdledford@redhat.com\u003e\nCc: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nCc: James Morris \u003cjames.l.morris@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "800179c9b8a1e796e441674776d11cd4c05d61d7", "tree": "5760992f4453c35b57b2686d8b8d5caee239b637", "parents": [ "3134f37e931d75931bdf6d4eacd82a3fd26eca7c" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Jul 25 17:29:07 2012 -0700" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sun Jul 29 21:37:58 2012 +0400" }, "message": "fs: add link restrictions\n\nThis adds symlink and hardlink restrictions to the Linux VFS.\n\nSymlinks:\n\nA long-standing class of security issues is the symlink-based\ntime-of-check-time-of-use race, most commonly seen in world-writable\ndirectories like /tmp. The common method of exploitation of this flaw\nis to cross privilege boundaries when following a given symlink (i.e. a\nroot process follows a symlink belonging to another user). For a likely\nincomplete list of hundreds of examples across the years, please see:\nhttp://cve.mitre.org/cgi-bin/cvekey.cgi?keyword\u003d/tmp\n\nThe solution is to permit symlinks to only be followed when outside\na sticky world-writable directory, or when the uid of the symlink and\nfollower match, or when the directory owner matches the symlink\u0027s owner.\n\nSome pointers to the history of earlier discussion that I could find:\n\n 1996 Aug, Zygo Blaxell\n http://marc.info/?l\u003dbugtraq\u0026m\u003d87602167419830\u0026w\u003d2\n 1996 Oct, Andrew Tridgell\n http://lkml.indiana.edu/hypermail/linux/kernel/9610.2/0086.html\n 1997 Dec, Albert D Cahalan\n http://lkml.org/lkml/1997/12/16/4\n 2005 Feb, Lorenzo Hernández García-Hierro\n http://lkml.indiana.edu/hypermail/linux/kernel/0502.0/1896.html\n 2010 May, Kees Cook\n https://lkml.org/lkml/2010/5/30/144\n\nPast objections and rebuttals could be summarized as:\n\n - Violates POSIX.\n - POSIX didn\u0027t consider this situation and it\u0027s not useful to follow\n a broken specification at the cost of security.\n - Might break unknown applications that use this feature.\n - Applications that break because of the change are easy to spot and\n fix. Applications that are vulnerable to symlink ToCToU by not having\n the change aren\u0027t. Additionally, no applications have yet been found\n that rely on this behavior.\n - Applications should just use mkstemp() or O_CREATE|O_EXCL.\n - True, but applications are not perfect, and new software is written\n all the time that makes these mistakes; blocking this flaw at the\n kernel is a single solution to the entire class of vulnerability.\n - This should live in the core VFS.\n - This should live in an LSM. (https://lkml.org/lkml/2010/5/31/135)\n - This should live in an LSM.\n - This should live in the core VFS. (https://lkml.org/lkml/2010/8/2/188)\n\nHardlinks:\n\nOn systems that have user-writable directories on the same partition\nas system files, a long-standing class of security issues is the\nhardlink-based time-of-check-time-of-use race, most commonly seen in\nworld-writable directories like /tmp. The common method of exploitation\nof this flaw is to cross privilege boundaries when following a given\nhardlink (i.e. a root process follows a hardlink created by another\nuser). Additionally, an issue exists where users can \"pin\" a potentially\nvulnerable setuid/setgid file so that an administrator will not actually\nupgrade a system fully.\n\nThe solution is to permit hardlinks to only be created when the user is\nalready the existing file\u0027s owner, or if they already have read/write\naccess to the existing file.\n\nMany Linux users are surprised when they learn they can link to files\nthey have no access to, so this change appears to follow the doctrine\nof \"least surprise\". Additionally, this change does not violate POSIX,\nwhich states \"the implementation may require that the calling process\nhas permission to access the existing file\"[1].\n\nThis change is known to break some implementations of the \"at\" daemon,\nthough the version used by Fedora and Ubuntu has been fixed[2] for\na while. Otherwise, the change has been undisruptive while in use in\nUbuntu for the last 1.5 years.\n\n[1] http://pubs.opengroup.org/onlinepubs/9699919799/functions/linkat.html\n[2] http://anonscm.debian.org/gitweb/?p\u003dcollab-maint/at.git;a\u003dcommitdiff;h\u003df4114656c3a6c6f6070e315ffdf940a49eda3279\n\nThis patch is based on the patches in Openwall and grsecurity, along with\nsuggestions from Al Viro. I have added a sysctl to enable the protected\nbehavior, and documentation.\n\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "620f6e8e855d6d447688a5f67a4e176944a084e8", "tree": "2be8f032eefc0d3bdaf9a045bd9263da4a2c6823", "parents": [ "6c216ec636f75d834461be15f83ec41a6759bd2b" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Apr 04 11:40:19 2012 -0700" }, "committer": { "name": "James Morris", "email": "james.l.morris@oracle.com", "time": "Thu Apr 05 14:51:43 2012 +1000" }, "message": "sysctl: fix write access to dmesg_restrict/kptr_restrict\n\nCommit bfdc0b4 adds code to restrict access to dmesg_restrict,\nhowever, it incorrectly alters kptr_restrict rather than\ndmesg_restrict.\n\nThe original patch from Richard Weinberger\n(https://lkml.org/lkml/2011/3/14/362) alters dmesg_restrict as\nexpected, and so the patch seems to have been misapplied.\n\nThis adds the CAP_SYS_ADMIN check to both dmesg_restrict and\nkptr_restrict, since both are sensitive.\n\nReported-by: Phillip Lougher \u003cplougher@redhat.com\u003e\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nAcked-by: Richard Weinberger \u003crichard@nod.at\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: James Morris \u003cjames.l.morris@oracle.com\u003e\n" }, { "commit": "532bfc851a7475fb6a36c1e953aa395798a7cca7", "tree": "a7892e5a31330dd59f31959efbe9fda1803784fd", "parents": [ "0195c00244dc2e9f522475868fa278c473ba7339", "8da00edc1069f01c34510fa405dc15d96c090a3f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 17:19:27 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 17:19:28 2012 -0700" }, "message": "Merge branch \u0027akpm\u0027 (Andrew\u0027s patch-bomb)\n\nMerge third batch of patches from Andrew Morton:\n - Some MM stragglers\n - core SMP library cleanups (on_each_cpu_mask)\n - Some IPI optimisations\n - kexec\n - kdump\n - IPMI\n - the radix-tree iterator work\n - various other misc bits.\n\n \"That\u0027ll do for -rc1. I still have ~10 patches for 3.4, will send\n those along when they\u0027ve baked a little more.\"\n\n* emailed from Andrew Morton \u003cakpm@linux-foundation.org\u003e: (35 commits)\n backlight: fix typo in tosa_lcd.c\n crc32: add help text for the algorithm select option\n mm: move hugepage test examples to tools/testing/selftests/vm\n mm: move slabinfo.c to tools/vm\n mm: move page-types.c from Documentation to tools/vm\n selftests/Makefile: make `run_tests\u0027 depend on `all\u0027\n selftests: launch individual selftests from the main Makefile\n radix-tree: use iterators in find_get_pages* functions\n radix-tree: rewrite gang lookup using iterator\n radix-tree: introduce bit-optimized iterator\n fs/proc/namespaces.c: prevent crash when ns_entries[] is empty\n nbd: rename the nbd_device variable from lo to nbd\n pidns: add reboot_pid_ns() to handle the reboot syscall\n sysctl: use bitmap library functions\n ipmi: use locks on watchdog timeout set on reboot\n ipmi: simplify locking\n ipmi: fix message handling during panics\n ipmi: use a tasklet for handling received messages\n ipmi: increase KCS timeouts\n ipmi: decrease the IPMI message transaction time in interrupt mode\n ...\n" }, { "commit": "5a04cca6c39cdd0b8c75b0628da634248f381b62", "tree": "3bb72a3fb05be731f19e97b48d1744783747f43b", "parents": [ "423a5bb49ec530ec8bbfc73fd2ded83da8e58684" ], "author": { "name": "Akinobu Mita", "email": "akinobu.mita@gmail.com", "time": "Wed Mar 28 14:42:50 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 17:14:36 2012 -0700" }, "message": "sysctl: use bitmap library functions\n\nUse bitmap_set() instead of using set_bit() for each bit. This conversion\nis valid because the bitmap is private in the function call and atomic\nbitops were unnecessary.\n\nThis also includes minor change.\n- Use bitmap_copy() for shorter typing\n\nSigned-off-by: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0195c00244dc2e9f522475868fa278c473ba7339", "tree": "f97ca98ae64ede2c33ad3de05ed7bbfa4f4495ed", "parents": [ "f21ce8f8447c8be8847dadcfdbcc76b0d7365fa5", "141124c02059eee9dbc5c86ea797b1ca888e77f7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 15:58:21 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 15:58:21 2012 -0700" }, "message": "Merge tag \u0027split-asm_system_h-for-linus-20120328\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-asm_system\n\nPull \"Disintegrate and delete asm/system.h\" from David Howells:\n \"Here are a bunch of patches to disintegrate asm/system.h into a set of\n separate bits to relieve the problem of circular inclusion\n dependencies.\n\n I\u0027ve built all the working defconfigs from all the arches that I can\n and made sure that they don\u0027t break.\n\n The reason for these patches is that I recently encountered a circular\n dependency problem that came about when I produced some patches to\n optimise get_order() by rewriting it to use ilog2().\n\n This uses bitops - and on the SH arch asm/bitops.h drags in\n asm-generic/get_order.h by a circuituous route involving asm/system.h.\n\n The main difficulty seems to be asm/system.h. It holds a number of\n low level bits with no/few dependencies that are commonly used (eg.\n memory barriers) and a number of bits with more dependencies that\n aren\u0027t used in many places (eg. switch_to()).\n\n These patches break asm/system.h up into the following core pieces:\n\n (1) asm/barrier.h\n\n Move memory barriers here. This already done for MIPS and Alpha.\n\n (2) asm/switch_to.h\n\n Move switch_to() and related stuff here.\n\n (3) asm/exec.h\n\n Move arch_align_stack() here. Other process execution related bits\n could perhaps go here from asm/processor.h.\n\n (4) asm/cmpxchg.h\n\n Move xchg() and cmpxchg() here as they\u0027re full word atomic ops and\n frequently used by atomic_xchg() and atomic_cmpxchg().\n\n (5) asm/bug.h\n\n Move die() and related bits.\n\n (6) asm/auxvec.h\n\n Move AT_VECTOR_SIZE_ARCH here.\n\n Other arch headers are created as needed on a per-arch basis.\"\n\nFixed up some conflicts from other header file cleanups and moving code\naround that has happened in the meantime, so David\u0027s testing is somewhat\nweakened by that. We\u0027ll find out anything that got broken and fix it..\n\n* tag \u0027split-asm_system_h-for-linus-20120328\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-asm_system: (38 commits)\n Delete all instances of asm/system.h\n Remove all #inclusions of asm/system.h\n Add #includes needed to permit the removal of asm/system.h\n Move all declarations of free_initmem() to linux/mm.h\n Disintegrate asm/system.h for OpenRISC\n Split arch_align_stack() out from asm-generic/system.h\n Split the switch_to() wrapper out of asm-generic/system.h\n Move the asm-generic/system.h xchg() implementation to asm-generic/cmpxchg.h\n Create asm-generic/barrier.h\n Make asm-generic/cmpxchg.h #include asm-generic/cmpxchg-local.h\n Disintegrate asm/system.h for Xtensa\n Disintegrate asm/system.h for Unicore32 [based on ver #3, changed by gxt]\n Disintegrate asm/system.h for Tile\n Disintegrate asm/system.h for Sparc\n Disintegrate asm/system.h for SH\n Disintegrate asm/system.h for Score\n Disintegrate asm/system.h for S390\n Disintegrate asm/system.h for PowerPC\n Disintegrate asm/system.h for PA-RISC\n Disintegrate asm/system.h for MN10300\n ...\n" }, { "commit": "9ffc93f203c18a70623f21950f1dd473c9ec48cd", "tree": "1eb3536ae183b0bfbf7f5152a6fe4f430ae881c2", "parents": [ "96f951edb1f1bdbbc99b0cd458f9808bb83d58ae" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Mar 28 18:30:03 2012 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Mar 28 18:30:03 2012 +0100" }, "message": "Remove all #inclusions of asm/system.h\n\nRemove all #inclusions of asm/system.h preparatory to splitting and killing\nit. Performed with the following command:\n\nperl -p -i -e \u0027s!^#\\s*include\\s*\u003casm/system[.]h\u003e.*\\n!!\u0027 `grep -Irl \u0027^#\\s*include\\s*\u003casm/system[.]h\u003e\u0027 *`\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\n" }, { "commit": "d550bbd40c0e10aefa05103dadbe0ae42e683707", "tree": "c1f32662f605f2b87c73901fbf66d12b2ca69b04", "parents": [ "e839ca528718e68cad32a307dc9aabf01ef3eb05" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Mar 28 18:30:03 2012 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Mar 28 18:30:03 2012 +0100" }, "message": "Disintegrate asm/system.h for Sparc\n\nDisintegrate asm/system.h for Sparc.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: sparclinux@vger.kernel.org\n" }, { "commit": "f1d38e423a697b7aa06e12d3ca4753bcc1aa3531", "tree": "1cbfd86070f724d5ffe53146d4c67edf14cccf98", "parents": [ "dae430c6f6e5d0b98c238c340a41a39e221e8940", "4e474a00d7ff746ed177ddae14fa8b2d4bad7a00" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 23 18:08:58 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 23 18:08:58 2012 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/sysctl\n\nPull sysctl updates from Eric Biederman:\n\n - Rewrite of sysctl for speed and clarity.\n\n Insert/remove/Lookup in sysctl are all now O(NlogN) operations, and\n are no longer bottlenecks in the process of adding and removing\n network devices.\n\n sysctl is now focused on being a filesystem instead of system call\n and the code can all be found in fs/proc/proc_sysctl.c. Hopefully\n this means the code is now approachable.\n\n Much thanks is owed to Lucian Grinjincu for keeping at this until\n something was found that was usable.\n\n - The recent proc_sys_poll oops found by the fuzzer during hibernation\n is fixed.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/sysctl: (36 commits)\n sysctl: protect poll() in entries that may go away\n sysctl: Don\u0027t call sysctl_follow_link unless we are a link.\n sysctl: Comments to make the code clearer.\n sysctl: Correct error return from get_subdir\n sysctl: An easier to read version of find_subdir\n sysctl: fix memset parameters in setup_sysctl_set()\n sysctl: remove an unused variable\n sysctl: Add register_sysctl for normal sysctl users\n sysctl: Index sysctl directories with rbtrees.\n sysctl: Make the header lists per directory.\n sysctl: Move sysctl_check_dups into insert_header\n sysctl: Modify __register_sysctl_paths to take a set instead of a root and an nsproxy\n sysctl: Replace root_list with links between sysctl_table_sets.\n sysctl: Add sysctl_print_dir and use it in get_subdir\n sysctl: Stop requiring explicit management of sysctl directories\n sysctl: Add a root pointer to ctl_table_set\n sysctl: Rewrite proc_sys_readdir in terms of first_entry and next_entry\n sysctl: Rewrite proc_sys_lookup introducing find_entry and lookup_entry.\n sysctl: Normalize the root_table data structure.\n sysctl: Factor out insert_header and erase_header\n ...\n" }, { "commit": "4040153087478993cbf0809f444400a3c808074c", "tree": "2dc7af85b0cf930f1656553bd38410b8c16601a6", "parents": [ "191c542442fdf53cc3c496c00be13367fd9cd42d" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Mon Feb 13 03:58:52 2012 +0000" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue Feb 14 10:45:42 2012 +1100" }, "message": "security: trim security.h\n\nTrim security.h\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "1f87f0b52b1d6581168cb80f86746bc4df918d01", "tree": "73576c0872c61c526ade225d6441a6610beb666b", "parents": [ "de4e83bd6b5e16d491ec068cd22801d5d063b07a" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Jan 06 04:07:15 2012 -0800" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue Jan 24 16:37:54 2012 -0800" }, "message": "sysctl: Move the implementation into fs/proc/proc_sysctl.c\n\nMove the core sysctl code from kernel/sysctl.c and kernel/sysctl_check.c\ninto fs/proc/proc_sysctl.c.\n\nCurrently sysctl maintenance is hampered by the sysctl implementation\nbeing split across 3 files with artificial layering between them.\nConsolidate the entire sysctl implementation into 1 file so that\nit is easier to see what is going on and hopefully allowing for\nsimpler maintenance.\n\nFor functions that are now only used in fs/proc/proc_sysctl.c remove\ntheir declarations from sysctl.h and make them static in fs/proc/proc_sysctl.c\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n" }, { "commit": "de4e83bd6b5e16d491ec068cd22801d5d063b07a", "tree": "d6b38f1284186abe5f20cfe996be9a4853c503b4", "parents": [ "0ce8974d504913a0f0ae2d97b20a5ac665431a41" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Jan 06 03:34:20 2012 -0800" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue Jan 24 16:37:54 2012 -0800" }, "message": "sysctl: Register the base sysctl table like any other sysctl table.\n\nSimplify the code by treating the base sysctl table like any other\nsysctl table and register it with register_sysctl_table.\n\nTo ensure this table is registered early enough to avoid problems\ncall sysctl_init from proc_sys_init.\n\nRename sysctl_net.c:sysctl_init() to net_sysctl_init() to avoid\nname conflicts now that kernel/sysctl.c:sysctl_init() is no longer\nstatic.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n" }, { "commit": "0ce8974d504913a0f0ae2d97b20a5ac665431a41", "tree": "50edf9b96066a3ca66143da742980dc04d72fb05", "parents": [ "36885d7b1121c779e4060d45472fe53a5b21e09f" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Jan 06 03:13:27 2012 -0800" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue Jan 24 16:37:54 2012 -0800" }, "message": "sysctl: Consolidate !CONFIG_SYSCTL handling\n\n- In sysctl.h move functions only available if CONFIG_SYSCL\n is defined inside of #ifdef CONFIG_SYSCTL\n\n- Move the stub function definitions for !CONFIG_SYSCTL\n into sysctl.h and make them static inlines.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n" }, { "commit": "55af77969fbd7a841838220ea2287432e0da8ae5", "tree": "f5f819e160799fb0477ec33c06aa8d022466687d", "parents": [ "37fe6a42b3433b79a159ceb06a94cd1ef00e279d" ], "author": { "name": "Mitsuo Hayasaka", "email": "mitsuo.hayasaka.hu@hitachi.com", "time": "Tue Nov 29 15:08:36 2011 +0900" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Dec 05 11:37:47 2011 +0100" }, "message": "x86: Panic on detection of stack overflow\n\nCurrently, messages are just output on the detection of stack\noverflow, which is not sufficient for systems that need a\nhigh reliability. This is because in general the overflow may\ncorrupt data, and the additional corruption may occur due to\nreading them unless systems stop.\n\nThis patch adds the sysctl parameter\nkernel.panic_on_stackoverflow and causes a panic when detecting\nthe overflows of kernel, IRQ and exception stacks except user\nstack according to the parameter. It is disabled by default.\n\nSigned-off-by: Mitsuo Hayasaka \u003cmitsuo.hayasaka.hu@hitachi.com\u003e\nCc: yrl.pp-manager.tt@hitachi.com\nCc: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nLink: http://lkml.kernel.org/r/20111129060836.11076.12323.stgit@ltc219.sdl.hitachi.co.jp\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "094803e0aab3fe75bbf8202a8f4b5280eaade375", "tree": "278528ca9245a767fcfcfa97d977bd5714c082fd", "parents": [ "32087d4eeca14b82660dab288b1d659963b954bd", "d8805e633e054c816c47cb6e727c81f156d9253d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 31 17:46:07 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 31 17:46:07 2011 -0700" }, "message": "Merge branch \u0027akpm\u0027 (Andrew\u0027s incoming)\n\nQuoth Andrew:\n\n - Most of MM. Still waiting for the poweroc guys to get off their\n butts and review some threaded hugepages patches.\n\n - alpha\n\n - vfs bits\n\n - drivers/misc\n\n - a few core kerenl tweaks\n\n - printk() features\n\n - MAINTAINERS updates\n\n - backlight merge\n\n - leds merge\n\n - various lib/ updates\n\n - checkpatch updates\n\n* akpm: (127 commits)\n epoll: fix spurious lockdep warnings\n checkpatch: add a --strict check for utf-8 in commit logs\n kernel.h/checkpatch: mark strict_strto\u003cfoo\u003e and simple_strto\u003cfoo\u003e as obsolete\n llist-return-whether-list-is-empty-before-adding-in-llist_add-fix\n wireless: at76c50x: follow rename pack_hex_byte to hex_byte_pack\n fat: follow rename pack_hex_byte() to hex_byte_pack()\n security: follow rename pack_hex_byte() to hex_byte_pack()\n kgdb: follow rename pack_hex_byte() to hex_byte_pack()\n lib: rename pack_hex_byte() to hex_byte_pack()\n lib/string.c: fix strim() semantics for strings that have only blanks\n lib/idr.c: fix comment for ida_get_new_above()\n lib/percpu_counter.c: enclose hotplug only variables in hotplug ifdef\n lib/bitmap.c: quiet sparse noise about address space\n lib/spinlock_debug.c: print owner on spinlock lockup\n lib/kstrtox: common code between kstrto*() and simple_strto*() functions\n drivers/leds/leds-lp5521.c: check if reset is successful\n leds: turn the blink_timer off before starting to blink\n leds: save the delay values after a successful call to blink_set()\n drivers/leds/leds-gpio.c: use gpio_get_value_cansleep() when initializing\n drivers/leds/leds-lm3530.c: add __devexit_p where needed\n ...\n" }, { "commit": "73efc0394e148d0e15583e13712637831f926720", "tree": "896c58a85f520bcccae4f16647cfb0f1ecd2c08e", "parents": [ "4ff819515b203f937cc6c8a0215a37a68d1ee71f" ], "author": { "name": "Dan Ballard", "email": "dan@mindstab.net", "time": "Mon Oct 31 17:11:20 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 31 17:30:53 2011 -0700" }, "message": "kernel/sysctl.c: add cap_last_cap to /proc/sys/kernel\n\nUserspace needs to know the highest valid capability of the running\nkernel, which right now cannot reliably be retrieved from the header files\nonly. The fact that this value cannot be determined properly right now\ncreates various problems for libraries compiled on newer header files\nwhich are run on older kernels. They assume capabilities are available\nwhich actually aren\u0027t. libcap-ng is one example. And we ran into the\nsame problem with systemd too.\n\nNow the capability is exported in /proc/sys/kernel/cap_last_cap.\n\n[akpm@linux-foundation.org: make cap_last_cap const, per Ulrich]\nSigned-off-by: Dan Ballard \u003cdan@mindstab.net\u003e\nCc: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Lennart Poettering \u003clennart@poettering.net\u003e\nCc: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nCc: Ulrich Drepper \u003cdrepper@akkadia.org\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "638ad34a8811119b32247b7722288ef8b90907d1", "tree": "9153571e8de8f9247bb54f9baddb140fd5fa5684", "parents": [ "e54aafa0c3bef84bfd39b4c713942ae7cdcfc58a" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Sun Oct 30 15:17:13 2011 +0100" }, "committer": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Sun Oct 30 15:16:46 2011 +0100" }, "message": "[S390] sparse: fix sparse warnings about missing prototypes\n\nAdd prototypes and includes for functions used in different modules.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "ec12cb7f31e28854efae7dd6f9544e0a66379040", "tree": "30a7293a4f9d566043f524bb4c43d4ae8b0560db", "parents": [ "a790de99599a29ad3f18667530cf4b9f4b7e3234" ], "author": { "name": "Paul Turner", "email": "pjt@google.com", "time": "Thu Jul 21 09:43:30 2011 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Aug 14 12:03:26 2011 +0200" }, "message": "sched: Accumulate per-cfs_rq cpu usage and charge against bandwidth\n\nAccount bandwidth usage on the cfs_rq level versus the task_groups to which\nthey belong. Whether we are tracking bandwidth on a given cfs_rq is maintained\nunder cfs_rq-\u003eruntime_enabled.\n\ncfs_rq\u0027s which belong to a bandwidth constrained task_group have their runtime\naccounted via the update_curr() path, which withdraws bandwidth from the global\npool as desired. Updates involving the global pool are currently protected\nunder cfs_bandwidth-\u003elock, local runtime is protected by rq-\u003elock.\n\nThis patch only assigns and tracks quota, no action is taken in the case that\ncfs_rq-\u003eruntime_used exceeds cfs_rq-\u003eruntime_assigned.\n\nSigned-off-by: Paul Turner \u003cpjt@google.com\u003e\nSigned-off-by: Nikhil Rao \u003cncrao@google.com\u003e\nSigned-off-by: Bharata B Rao \u003cbharata@linux.vnet.ibm.com\u003e\nReviewed-by: Hidetoshi Seto \u003cseto.hidetoshi@jp.fujitsu.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLink: http://lkml.kernel.org/r/20110721184757.179386821@google.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "a95cded32de3deae13af34715200532e6823cc9f", "tree": "b6085f03d7fac7993a655e3796b4286c4db96694", "parents": [ "22a3c7d188c2b7bfc8e949bf9fad215c094ba78b" ], "author": { "name": "Paul E. McKenney", "email": "paulmck@linux.vnet.ibm.com", "time": "Sun May 01 23:21:00 2011 -0700" }, "committer": { "name": "Paul E. McKenney", "email": "paulmck@linux.vnet.ibm.com", "time": "Wed Jul 20 14:10:18 2011 -0700" }, "message": "sysctl,rcu: Convert call_rcu(free_head) to kfree\n\nThe RCU callback free_head just calls kfree(), so we can use kfree_rcu()\ninstead of call_rcu().\n\nSigned-off-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Josh Triplett \u003cjosh@joshtriplett.org\u003e\n" }, { "commit": "aa4a221875873d2a1f9656cb7fd7e545e952b4fa", "tree": "44f66345c06c94cfc95d17b337d45e09917e475d", "parents": [ "710054ba25c0d1f8f41c22ce13ba336503fb5318" ], "author": { "name": "Vince Weaver", "email": "vweaver1@eecs.utk.edu", "time": "Fri Jun 03 17:54:40 2011 -0400" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Jun 04 12:22:04 2011 +0200" }, "message": "perf: Comment /proc/sys/kernel/perf_event_paranoid to be part of user ABI\n\nTurns out that distro packages use this file as an indicator of\nthe perf event subsystem - this is easier to check for from scripts\nthan the existence of the system call.\n\nThis is easy enough to keep around for the kernel, so add a\ncomment to make sure it stays so.\n\nSigned-off-by: Vince Weaver \u003cvweaver1@eecs.utk.edu\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: paulus@samba.org\nCc: acme@redhat.com\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/alpine.DEB.2.00.1106031751170.29381@cl320.eecs.utk.edu\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "0798b1dbfbd9ff2a370c5968c5f0621ef0075fe0", "tree": "c7f61ab9683786a070da0933b9981fc74a4d865f", "parents": [ "ad363e0916423b2e6cdfcdc30ae707ec709f0a65", "6738d3210aabe3016a1b03cd98a7fc479c229197" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 15:35:32 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 15:35:32 2011 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile: (26 commits)\n arch/tile: prefer \"tilepro\" as the name of the 32-bit architecture\n compat: include aio_abi.h for aio_context_t\n arch/tile: cleanups for tilegx compat mode\n arch/tile: allocate PCI IRQs later in boot\n arch/tile: support signal \"exception-trace\" hook\n arch/tile: use better definitions of xchg() and cmpxchg()\n include/linux/compat.h: coding-style fixes\n tile: add an RTC driver for the Tilera hypervisor\n arch/tile: finish enabling support for TILE-Gx 64-bit chip\n compat: fixes to allow working with tile arch\n arch/tile: update defconfig file to something more useful\n tile: do_hardwall_trap: do not play with task-\u003esighand\n tile: replace mm-\u003ecpu_vm_mask with mm_cpumask()\n tile,mn10300: add device parameter to dma_cache_sync()\n audit: support the \"standard\" \u003casm-generic/unistd.h\u003e\n arch/tile: clarify flush_buffer()/finv_buffer() function names\n arch/tile: kernel-related cleanups from removing static page size\n arch/tile: various header improvements for building drivers\n arch/tile: disable GX prefetcher during cache flush\n arch/tile: tolerate disabling CONFIG_BLK_DEV_INITRD\n ...\n" }, { "commit": "434d42cfd05a7cc452457a81d2029540cba12150", "tree": "3a6b9b7f9ff2e1b7409dd66c15242b2a75aa4422", "parents": [ "d762f4383100c2a87b1a3f2d678cd3b5425655b4", "12a5a2621b1ee14d32beca35304d7c6076a58815" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue May 24 22:55:24 2011 +1000" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue May 24 22:55:24 2011 +1000" }, "message": "Merge branch \u0027next\u0027 into for-linus\n" }, { "commit": "586692a5a5fc5740c8a46abc0f2365495c2d7c5f", "tree": "bc08228e67a968d83691c9efc5ea1feda9e6f98b", "parents": [ "e04ab2bc41b35c0cb6cdb07c8443f91aa738cf78" ], "author": { "name": "Mandeep Singh Baines", "email": "msb@chromium.org", "time": "Sun May 22 22:10:22 2011 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 23 11:58:59 2011 +0200" }, "message": "watchdog: Disable watchdog when thresh is zero\n\nThis restores the previous behavior of softlock_thresh.\n\nCurrently, setting watchdog_thresh to zero causes the watchdog\nkthreads to consume a lot of CPU.\n\nIn addition, the logic of proc_dowatchdog_thresh and\nproc_dowatchdog_enabled has been factored into proc_dowatchdog.\n\nSigned-off-by: Mandeep Singh Baines \u003cmsb@chromium.org\u003e\nCc: Marcin Slusarz \u003cmarcin.slusarz@gmail.com\u003e\nCc: Don Zickus \u003cdzickus@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nLink: http://lkml.kernel.org/r/1306127423-3347-3-git-send-email-msb@chromium.org\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nLKML-Reference: \u003c20110517071018.GE22305@elte.hu\u003e\n" }, { "commit": "571d76acdab95876aeff869ab6449f826c23aa43", "tree": "b52ceacfa83b1ab4c5a6950007ce8be03cec192e", "parents": [ "8aaf1dda42576b0f8dffb004065baa806f4df9b6" ], "author": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Mon May 16 14:23:44 2011 -0400" }, "committer": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Thu May 19 22:55:59 2011 -0400" }, "message": "arch/tile: support signal \"exception-trace\" hook\n\nThis change adds support for /proc/sys/debug/exception-trace to tile.\nLike x86 and sparc, by default it is set to \"1\", generating a one-line\nprintk whenever a user process crashes. By setting it to \"2\", we get\na much more complete userspace diagnostic at crash time, including\na user-space backtrace, register dump, and memory dump around the\naddress of the crash.\n\nSome vestiges of the Tilera-internal version of this support are\nremoved with this patch (the show_crashinfo variable and the\narch_coredump_signal function). We retain a \"crashinfo\" boot parameter\nwhich allows you to set the boot-time value of exception-trace.\n\nSigned-off-by: Chris Metcalf \u003ccmetcalf@tilera.com\u003e\n" }, { "commit": "17f60a7da150fdd0cfb9756f86a262daa72c835f", "tree": "1c5ece75d66bed0766de861cde6eb18ee7ea4cf2", "parents": [ "cfc64fd91fabed099a4c3df58559f4b7efe9bcce" ], "author": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Fri Apr 01 17:07:50 2011 -0400" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Mon Apr 04 10:31:04 2011 +1000" }, "message": "capabilites: allow the application of capability limits to usermode helpers\n\nThere is no way to limit the capabilities of usermodehelpers. This problem\nreared its head recently when someone complained that any user with\ncap_net_admin was able to load arbitrary kernel modules, even though the user\ndidn\u0027t have cap_sys_module. The reason is because the actual load is done by\na usermode helper and those always have the full cap set. This patch addes new\nsysctls which allow us to bound the permissions of usermode helpers.\n\n/proc/sys/kernel/usermodehelper/bset\n/proc/sys/kernel/usermodehelper/inheritable\n\nYou must have CAP_SYS_MODULE and CAP_SETPCAP to change these (changes are\n\u0026\u003d ONLY). When the kernel launches a usermodehelper it will do so with these\nas the bset and pI.\n\n-v2:\tmake globals static\n\tcreate spinlock to protect globals\n\n-v3:\trequire both CAP_SETPCAP and CAP_SYS_MODULE\n-v4:\tfix the typo s/CAP_SET_PCAP/CAP_SETPCAP/ because I didn\u0027t commit\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\nNo-objection-from: Serge E. Hallyn \u003cserge.hallyn@canonical.com\u003e\nAcked-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Serge E. Hallyn \u003cserge.hallyn@canonical.com\u003e\nAcked-by: Andrew G. Morgan \u003cmorgan@kernel.org\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "bfdc0b497faa82a0ba2f9dddcf109231dd519fcc", "tree": "932897262447dacb7158b81209748a295d93e20b", "parents": [ "cb16e95fa2996743a6e80a665ed2ed0590bd38cf" ], "author": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Wed Mar 23 16:43:11 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 23 19:46:54 2011 -0700" }, "message": "sysctl: restrict write access to dmesg_restrict\n\nWhen dmesg_restrict is set to 1 CAP_SYS_ADMIN is needed to read the kernel\nring buffer. But a root user without CAP_SYS_ADMIN is able to reset\ndmesg_restrict to 0.\n\nThis is an issue when e.g. LXC (Linux Containers) are used and complete\nuser space is running without CAP_SYS_ADMIN. A unprivileged and jailed\nroot user can bypass the dmesg_restrict protection.\n\nWith this patch writing to dmesg_restrict is only allowed when root has\nCAP_SYS_ADMIN.\n\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\nAcked-by: Dan Rosenberg \u003cdrosenberg@vsecurity.com\u003e\nAcked-by: Serge E. Hallyn \u003cserge@hallyn.com\u003e\nCc: Eric Paris \u003ceparis@redhat.com\u003e\nCc: Kees Cook \u003ckees.cook@canonical.com\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nCc: Eugene Teo \u003ceugeneteo@kernel.org\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "cb16e95fa2996743a6e80a665ed2ed0590bd38cf", "tree": "7128a7d03a94a68dd9550f33ac66876bf6854ad0", "parents": [ "256c53a65128cbc8a766b1503f3f25a52a8d07cb" ], "author": { "name": "Petr Holasek", "email": "pholasek@redhat.com", "time": "Wed Mar 23 16:43:09 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 23 19:46:51 2011 -0700" }, "message": "sysctl: add some missing input constraint checks\n\nAdd boundaries of allowed input ranges for: dirty_expire_centisecs,\ndrop_caches, overcommit_memory, page-cluster and panic_on_oom.\n\nSigned-off-by: Petr Holasek \u003cpholasek@redhat.com\u003e\nAcked-by: Dave Young \u003chidave.darkstar@gmail.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0f6e0e8448a16d8d22119ce91d8dd24b44865b51", "tree": "7c295c02db035fc6a0b867465911a2bc9dc6b1ef", "parents": [ "0d2ecee2bdb2a19d04bc5cefac0f86e790f1aad4", "a002951c97ff8da49938c982a4c236bf2fafdc9f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 16 09:15:43 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 16 09:15:43 2011 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6: (33 commits)\n AppArmor: kill unused macros in lsm.c\n AppArmor: cleanup generated files correctly\n KEYS: Add an iovec version of KEYCTL_INSTANTIATE\n KEYS: Add a new keyctl op to reject a key with a specified error code\n KEYS: Add a key type op to permit the key description to be vetted\n KEYS: Add an RCU payload dereference macro\n AppArmor: Cleanup make file to remove cruft and make it easier to read\n SELinux: implement the new sb_remount LSM hook\n LSM: Pass -o remount options to the LSM\n SELinux: Compute SID for the newly created socket\n SELinux: Socket retains creator role and MLS attribute\n SELinux: Auto-generate security_is_socket_class\n TOMOYO: Fix memory leak upon file open.\n Revert \"selinux: simplify ioctl checking\"\n selinux: drop unused packet flow permissions\n selinux: Fix packet forwarding checks on postrouting\n selinux: Fix wrong checks for selinux_policycap_netpeer\n selinux: Fix check for xfrm selinux context algorithm\n ima: remove unnecessary call to ima_must_measure\n IMA: remove IMA imbalance checking\n ...\n" }, { "commit": "9620639b7ea3843983f4ced8b4c81eb4d8974838", "tree": "54266fac3bcf89e61ae06c7d36ca708df6e0ea33", "parents": [ "a926021cb1f8a99a275eaf6eb546102e9469dc59", "6d1cafd8b56ea726c10a5a104de57cc3ed8fa953" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:37:30 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:37:30 2011 -0700" }, "message": "Merge branch \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (26 commits)\n sched: Resched proper CPU on yield_to()\n sched: Allow users with sufficient RLIMIT_NICE to change from SCHED_IDLE policy\n sched: Allow SCHED_BATCH to preempt SCHED_IDLE tasks\n sched: Clean up the IRQ_TIME_ACCOUNTING code\n sched: Add #ifdef around irq time accounting functions\n sched, autogroup: Stop claiming ownership of the root task group\n sched, autogroup: Stop going ahead if autogroup is disabled\n sched, autogroup, sysctl: Use proc_dointvec_minmax() instead\n sched: Fix the group_imb logic\n sched: Clean up some f_b_g() comments\n sched: Clean up remnants of sd_idle\n sched: Wholesale removal of sd_idle logic\n sched: Add yield_to(task, preempt) functionality\n sched: Use a buddy to implement yield_task_fair()\n sched: Limit the scope of clear_buddies\n sched: Check the right -\u003enr_running in yield_task_fair()\n sched: Avoid expensive initial update_cfs_load(), on UP too\n sched: Fix switch_from_fair()\n sched: Simplify the idle scheduling class\n softirqs: Account ksoftirqd time as cpustat softirq\n ...\n" }, { "commit": "a926021cb1f8a99a275eaf6eb546102e9469dc59", "tree": "c6d0300cd4b1a1fd658708476db4577b68b4de31", "parents": [ "0586bed3e8563c2eb89bc7256e30ce633ae06cfb", "5e814dd597c42daeb8d2a276e64a6ec986ad0e2a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:31:30 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:31:30 2011 -0700" }, "message": "Merge branch \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (184 commits)\n perf probe: Clean up probe_point_lazy_walker() return value\n tracing: Fix irqoff selftest expanding max buffer\n tracing: Align 4 byte ints together in struct tracer\n tracing: Export trace_set_clr_event()\n tracing: Explain about unstable clock on resume with ring buffer warning\n ftrace/graph: Trace function entry before updating index\n ftrace: Add .ref.text as one of the safe areas to trace\n tracing: Adjust conditional expression latency formatting.\n tracing: Fix event alignment: skb:kfree_skb\n tracing: Fix event alignment: mce:mce_record\n tracing: Fix event alignment: kvm:kvm_hv_hypercall\n tracing: Fix event alignment: module:module_request\n tracing: Fix event alignment: ftrace:context_switch and ftrace:wakeup\n tracing: Remove lock_depth from event entry\n perf header: Stop using \u0027self\u0027\n perf session: Use evlist/evsel for managing perf.data attributes\n perf top: Don\u0027t let events to eat up whole header line\n perf top: Fix events overflow in top command\n ring-buffer: Remove unused #include \u003clinux/trace_irq.h\u003e\n tracing: Add an \u0027overwrite\u0027 trace_option.\n ...\n" }, { "commit": "a002951c97ff8da49938c982a4c236bf2fafdc9f", "tree": "d43e7885ea7376df0a47a0fc8ceca66dc5bfa357", "parents": [ "521cb40b0c44418a4fd36dc633f575813d59a43d", "c151694b2c48d956ac8c8c59c6927f89cc29ef70" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Wed Mar 16 09:41:17 2011 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Wed Mar 16 09:41:17 2011 +1100" }, "message": "Merge branch \u0027next\u0027 into for-linus\n" }, { "commit": "dfef6dcd35cb4a251f6322ca9b2c06f0bb1aa1f4", "tree": "65e8a25d4ed913658db35c4b97ab0a021c2124eb", "parents": [ "1858efd471624ecb37e6b5462cab8076f47d1cee" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 08 01:25:28 2011 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 08 02:22:27 2011 -0500" }, "message": "unfuck proc_sysctl -\u003ed_compare()\n\na) struct inode is not going to be freed under -\u003ed_compare();\nhowever, the thing PROC_I(inode)-\u003esysctl points to just might.\nFortunately, it\u0027s enough to make freeing that sucker delayed,\nprovided that we don\u0027t step on its -\u003eunregistering, clear\nthe pointer to it in PROC_I(inode) before dropping the reference\nand check if it\u0027s NULL in -\u003ed_compare().\n\nb) I\u0027m not sure that we *can* walk into NULL inode here (we recheck\ndentry-\u003eseq between verifying that it\u0027s still hashed / fetching\ndentry-\u003ed_inode and passing it to -\u003ed_compare() and there\u0027s no\nnegative hashed dentries in /proc/sys/*), but if we can walk into\nthat, we really should not have -\u003ed_compare() return 0 on it!\nSaid that, I really suspect that this check can be simply killed.\nNick?\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "fe3fa43039d47ee4e22caf460b79b62a14937f79", "tree": "9eab8d00f1227b9fe0959f32a62d892ed35803ba", "parents": [ "ee009e4a0d4555ed522a631bae9896399674f064", "026eb167ae77244458fa4b4b9fc171209c079ba7" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue Mar 08 11:38:10 2011 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue Mar 08 11:38:10 2011 +1100" }, "message": "Merge branch \u0027master\u0027 of git://git.infradead.org/users/eparis/selinux into next\n" }, { "commit": "1747b21fecbfb63fbf6b9624e8b92707960d5a97", "tree": "5c318fb7d8426f760f5cf3e3f1c256fb6b9fe1a8", "parents": [ "866ab43efd325fae8889ea77a744d03f2b957e38" ], "author": { "name": "Yong Zhang", "email": "yong.zhang0@gmail.com", "time": "Sun Feb 20 15:08:12 2011 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 23 11:33:58 2011 +0100" }, "message": "sched, autogroup, sysctl: Use proc_dointvec_minmax() instead\n\nsched_autogroup_enabled has min/max value, proc_dointvec_minmax() is\nbe used for this case.\n\nSigned-off-by: Yong Zhang \u003cyong.zhang0@gmail.com\u003e\nCc: Mike Galbraith \u003cefault@gmx.de\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLKML-Reference: \u003c1298185696-4403-2-git-send-email-yong.zhang0@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "48fa4b8ecf683f5e411303553da9e186e8b8406e", "tree": "76cec4aa8cca18a44b1bd6e5be68c462a0e66d5f", "parents": [ "d95f412200652694e63e64bfd49f0ae274a54479", "85e2efbb1db9a18d218006706d6e4fbeb0216213" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 16 13:31:51 2011 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 16 13:31:55 2011 +0100" }, "message": "Merge commit \u0027v2.6.38-rc5\u0027 into sched/core\n\nMerge reason: Pick up upstream fixes.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "163ec4354a5135c6c38c3f4a9b46a31900ebdf48", "tree": "59063e726453ace397c66d95cab09ac43265be41", "parents": [ "4979d2729af22f6ce8faa325fc60a85a2c2daa02" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 16 11:22:34 2011 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 16 13:30:55 2011 +0100" }, "message": "perf: Optimize throttling code\n\nBy pre-computing the maximum number of samples per tick we can avoid a\nmultiplication and a conditional since MAX_INTERRUPTS \u003e\nmax_samples_per_tick.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLKML-Reference: \u003cnew-submission\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "ac53db596cc08ecb8040cfb6f71ae40c6f2041c4", "tree": "c263b558772213530532026af6fa9e34a8f88375", "parents": [ "2c13c919d9e9a3db9896143a501f83dcbbe1ced4" ], "author": { "name": "Rik van Riel", "email": "riel@redhat.com", "time": "Tue Feb 01 09:51:03 2011 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Feb 03 14:20:33 2011 +0100" }, "message": "sched: Use a buddy to implement yield_task_fair()\n\nUse the buddy mechanism to implement yield_task_fair. This\nallows us to skip onto the next highest priority se at every\nlevel in the CFS tree, unless doing so would introduce gross\nunfairness in CPU time distribution.\n\nWe order the buddy selection in pick_next_entity to check\nyield first, then last, then next. We need next to be able\nto override yield, because it is possible for the \"next\" and\n\"yield\" task to be different processen in the same sub-tree\nof the CFS tree. When they are, we need to go into that\nsub-tree regardless of the \"yield\" hint, and pick the correct\nentity once we get to the right level.\n\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLKML-Reference: \u003c20110201095103.3a79e92a@annuminas.surriel.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4916ca401e3051dad326ddd69765bd0e3f32fb9b", "tree": "593778babcd691a498a909a5eaf462f29d241cf6", "parents": [ "8e6c96935fcc1ed3dbebc96fddfef3f2f2395afc" ], "author": { "name": "Lucian Adrian Grijincu", "email": "lucian.grijincu@gmail.com", "time": "Tue Feb 01 18:44:56 2011 +0200" }, "committer": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Tue Feb 01 11:54:02 2011 -0500" }, "message": "security: remove unused security_sysctl hook\n\nThe only user for this hook was selinux. sysctl routes every call\nthrough /proc/sys/. Selinux and other security modules use the file\nsystem checks for sysctl too, so no need for this hook any more.\n\nSigned-off-by: Lucian Adrian Grijincu \u003clucian.grijincu@gmail.com\u003e\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\n" }, { "commit": "6fb1b304255efc5c4c93874ac8c066272e257e28", "tree": "67b4193e20d3a5470f56b26d912ed791dba20f13", "parents": [ "ac751efa6a0d70f2c9daef5c7e3a92270f5c2dff", "409550f2902470f0387fe40a7db441526e16b2c0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jan 26 16:31:44 2011 +1000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jan 26 16:31:44 2011 +1000" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:\n Input: wacom - pass touch resolution to clients through input_absinfo\n Input: wacom - add 2 Bamboo Pen and touch models\n Input: sysrq - ensure sysrq_enabled and __sysrq_enabled are consistent\n Input: sparse-keymap - fix KEY_VSW handling in sparse_keymap_setup\n Input: tegra-kbc - add tegra keyboard driver\n Input: gpio_keys - switch to using request_any_context_irq\n Input: serio - allow registered drivers to get status flag\n Input: ct82710c - return proper error code for ct82c710_open\n Input: bu21013_ts - added regulator support\n Input: bu21013_ts - remove duplicate resolution parameters\n Input: tnetv107x-ts - don\u0027t treat NULL clk as an error\n Input: tnetv107x-keypad - don\u0027t treat NULL clk as an error\n\nFix up trivial conflicts in drivers/input/keyboard/Makefile due to\nadditions of tc3589x/Tegra drivers\n" }, { "commit": "8c6a98b22b750c9eb52653ba643faa17db8d3881", "tree": "682d083e544bf4b420b6972941c9f3e7493e77ef", "parents": [ "cb1b145929b5b7e1bdc5f76dd2905df9f89c038e" ], "author": { "name": "Andy Whitcroft", "email": "apw@canonical.com", "time": "Mon Jan 24 09:31:38 2011 -0800" }, "committer": { "name": "Dmitry Torokhov", "email": "dmitry.torokhov@gmail.com", "time": "Mon Jan 24 09:33:36 2011 -0800" }, "message": "Input: sysrq - ensure sysrq_enabled and __sysrq_enabled are consistent\n\nCurrently sysrq_enabled and __sysrq_enabled are initialised separately\nand inconsistently, leading to sysrq being actually enabled by reported\nas not enabled in sysfs. The first change to the sysfs configurable\nsynchronises these two:\n\n static int __read_mostly sysrq_enabled \u003d 1;\n static int __sysrq_enabled;\n\nAdd a common define to carry the default for these preventing them becoming\nout of sync again. Default this to 1 to mirror previous behaviour.\n\nSigned-off-by: Andy Whitcroft \u003capw@canonical.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Dmitry Torokhov \u003cdtor@mail.ru\u003e\n" }, { "commit": "e020e742e5dbd8c44d31706995dc13ddc732e274", "tree": "7a92c6bee050f27dda16dcc1eb3e7c9cf23fca07", "parents": [ "556105000334cb440636ef61b862d22b03c24f70" ], "author": { "name": "Jovi Zhang", "email": "bookjovi@gmail.com", "time": "Wed Jan 12 17:00:45 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 08:03:18 2011 -0800" }, "message": "sysctl: remove obsolete comments\n\nctl_unnumbered.txt have been removed in Documentation directory so just\nalso remove this invalid comments\n\n[akpm@linux-foundation.org: fix Documentation/sysctl/00-INDEX, per Dave]\nSigned-off-by: Jovi Zhang \u003cbookjovi@gmail.com\u003e\nCc: Dave Young \u003chidave.darkstar@gmail.com\u003e\nAcked-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "556105000334cb440636ef61b862d22b03c24f70", "tree": "7b5b8d859c4e6950a67e3785c77612b6f1773a00", "parents": [ "12a4dc43911785f51a596f771ae0701b18d436f1" ], "author": { "name": "Jovi Zhang", "email": "bookjovi@gmail.com", "time": "Wed Jan 12 17:00:45 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 08:03:18 2011 -0800" }, "message": "sysctl: fix #ifdef guard comment\n\nSigned-off-by: Jovi Zhang \u003cbookjovi@gmail.com\u003e\nAcked-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ], "next": "455cd5ab305c90ffc422dd2e0fb634730942b257" }