)]}' { "log": [ { "commit": "d1748302f70be7469809809283fe164156a34231", "tree": "c81137f4126f2fe1451c28441415c1cd0fa0f7cd", "parents": [ "29c158e81c733ac7d6a75c5ee929f34fb9f92983" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Tue Aug 23 15:29:42 2011 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Thu Sep 08 11:10:56 2011 +0200" }, "message": "clockevents: Make minimum delay adjustments configurable\n\nThe automatic increase of the min_delta_ns of a clockevents device\nshould be done in the clockevents code as the minimum delay is an\nattribute of the clockevents device.\n\nIn addition not all architectures want the automatic adjustment, on a\nmassively virtualized system it can happen that the programming of a\nclock event fails several times in a row because the virtual cpu has\nbeen rescheduled quickly enough. In that case the minimum delay will\nerroneously be increased with no way back. The new config symbol\nGENERIC_CLOCKEVENTS_MIN_ADJUST is used to enable the automatic\nadjustment. The config option is selected only for x86.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nLink: http://lkml.kernel.org/r/20110823133142.494157493@de.ibm.com\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "420c1c572d4ceaa2f37b6311b7017ac6cf049fe2", "tree": "df04e6b4b756b7a46d9887462d54a3ad0e1f91d5", "parents": [ "9620639b7ea3843983f4ced8b4c81eb4d8974838", "6e6823d17b157f185be09f4c70181299f9273f0b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:53:35 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 15 18:53:35 2011 -0700" }, "message": "Merge branch \u0027timers-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027timers-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (62 commits)\n posix-clocks: Check write permissions in posix syscalls\n hrtimer: Remove empty hrtimer_init_hres_timer()\n hrtimer: Update hrtimer-\u003estate documentation\n hrtimer: Update base[CLOCK_BOOTTIME].offset correctly\n timers: Export CLOCK_BOOTTIME via the posix timers interface\n timers: Add CLOCK_BOOTTIME hrtimer base\n time: Extend get_xtime_and_monotonic_offset() to also return sleep\n time: Introduce get_monotonic_boottime and ktime_get_boottime\n hrtimers: extend hrtimer base code to handle more then 2 clockids\n ntp: Remove redundant and incorrect parameter check\n mn10300: Switch do_timer() to xtimer_update()\n posix clocks: Introduce dynamic clocks\n posix-timers: Cleanup namespace\n posix-timers: Add support for fd based clocks\n x86: Add clock_adjtime for x86\n posix-timers: Introduce a syscall for clock tuning.\n time: Splitout compat timex accessors\n ntp: Add ADJ_SETOFFSET mode bit\n time: Introduce timekeeping_inject_offset\n posix-timer: Update comment\n ...\n\nFix up new system-call-related conflicts in\n\tarch/x86/ia32/ia32entry.S\n\tarch/x86/include/asm/unistd_32.h\n\tarch/x86/include/asm/unistd_64.h\n\tarch/x86/kernel/syscall_table_32.S\n(name_to_handle_at()/open_by_handle_at() vs clock_adjtime()), and some\ndue to movement of get_jiffies_64() in:\n\tkernel/time.c\n" }, { "commit": "3a142a0672b48a853f00af61f184c7341ac9c99d", "tree": "e84f2647dc2ae71f2e54f2256ebce87624c11d97", "parents": [ "493f3358cb289ccf716c5a14fa5bb52ab75943e5" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Feb 25 22:34:23 2011 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sat Feb 26 09:45:28 2011 +0100" }, "message": "clockevents: Prevent oneshot mode when broadcast device is periodic\n\nWhen the per cpu timer is marked CLOCK_EVT_FEAT_C3STOP, then we only\ncan switch into oneshot mode, when the backup broadcast device\nsupports oneshot mode as well. Otherwise we would try to switch the\nbroadcast device into an unsupported mode unconditionally. This went\nunnoticed so far as the current available broadcast devices support\noneshot mode. Seth unearthed this problem while debugging and working\naround an hpet related BIOS wreckage.\n\nAdd the necessary check to tick_is_oneshot_available().\n\nReported-and-tested-by: Seth Forshee \u003cseth.forshee@canonical.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nLKML-Reference: \u003calpine.LFD.2.00.1102252231200.2701@localhost6.localdomain6\u003e\nCc: stable@kernel.org # .21 -\u003e\n" }, { "commit": "e2830b5c1b2b2217894370a3b95af87d4a958401", "tree": "9eec3de82a30fffc32beae7826e4a356220c52f7", "parents": [ "d12b0e24c56c6fb2398609f26858e5278d688840" ], "author": { "name": "Torben Hohn", "email": "torbenh@gmx.de", "time": "Thu Jan 27 16:00:32 2011 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Mon Jan 31 19:26:50 2011 +0100" }, "message": "time: Make do_timer() and xtime_lock local to kernel/time/\n\nAll callers of do_timer() are converted to xtime_update(). The only\nusers of xtime_lock are in kernel/time/. Make both local to\nkernel/time/ and remove them from the global header files.\n\n[ tglx: Reuse tick-internal.h instead of creating another local header\n \tfile. Massaged changelog ]\n\nSigned-off-by: Torben Hohn \u003ctorbenh@gmx.de\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: johnstul@us.ibm.com\nCc: yong.zhang0@gmail.com\nCc: hch@infradead.org\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "909ea96468096b07fbb41aaf69be060d92bd9271", "tree": "a7e015edd96b5f674874fe78cdd889769e130a2a", "parents": [ "780f36d8b3fa9572f731d4fb85067b2e45e6f993" ], "author": { "name": "Christoph Lameter", "email": "cl@linux.com", "time": "Wed Dec 08 16:22:55 2010 +0100" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Dec 17 15:07:19 2010 +0100" }, "message": "core: Replace __get_cpu_var with __this_cpu_read if not used for an address.\n\n__get_cpu_var() can be replaced with this_cpu_read and will then use a\nsingle read instruction with implied address calculation to access the\ncorrect per cpu instance.\n\nHowever, the address of a per cpu variable passed to __this_cpu_read()\ncannot be determined (since it\u0027s an implied address conversion through\nsegment prefixes). Therefore apply this only to uses of __get_cpu_var\nwhere the address of the variable is not used.\n\nCc: Pekka Enberg \u003cpenberg@cs.helsinki.fi\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: H. Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Christoph Lameter \u003ccl@linux.com\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "b5f91da0a6973bb6f9ff3b91b0e92c0773a458f3", "tree": "911ad0eeb484ce9309504fcf5a031403ad47ce84", "parents": [ "d192c47f25daa98996c7eae543d8a27e41247ec2" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Dec 08 12:40:31 2009 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Mon Dec 14 23:55:34 2009 +0100" }, "message": "clockevents: Convert to raw_spinlock\n\nConvert locks which cannot be sleeping locks in preempt-rt to\nraw_spinlocks.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d192c47f25daa98996c7eae543d8a27e41247ec2", "tree": "7ad7970ffbf0fa25bab4fc9eb20f6a6fdbb30bb1", "parents": [ "aef9cb05247df3d7615773737beb4f83d78577bb" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Dec 08 12:49:26 2009 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Mon Dec 14 23:55:34 2009 +0100" }, "message": "clockevents: Make tick_device_lock static\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "74a03b69d1b5ce00a568e142ca97e76b7f5239c6", "tree": "02bdea43ae6d528dcee97c00c3a8651f8841411a", "parents": [ "091438dd5668396328a3419abcbc6591159eb8d1" ], "author": { "name": "john stultz", "email": "johnstul@us.ibm.com", "time": "Fri May 01 13:10:25 2009 -0700" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sat May 02 10:22:27 2009 +0200" }, "message": "clockevents: prevent endless loop in tick_handle_periodic()\n\ntick_handle_periodic() can lock up hard when a one shot clock event\ndevice is used in combination with jiffies clocksource.\n\nAvoid an endless loop issue by requiring that a highres valid\nclocksource be installed before we call tick_periodic() in a loop when\nusing ONESHOT mode. The result is we will only increment jiffies once\nper interrupt until a continuous hardware clocksource is available.\n\nWithout this, we can run into a endless loop, where each cycle through\nthe loop, jiffies is updated which increments time by tick_period or\nmore (due to clock steering), which can cause the event programming to\nthink the next event was before the newly incremented time and fail\ncausing tick_periodic() to be called again and the whole process loops\nforever.\n\n[ Impact: prevent hard lock up ]\n\nSigned-off-by: John Stultz \u003cjohnstul@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: stable@kernel.org\n" }, { "commit": "94df7de0289bc2df3d6e85cd2ece52bf42682f45", "tree": "b3f614b015adfb9574959687bb9b7ac4c884e23a", "parents": [ "7f22391cbe82a80a9f891d8bd10fc28ff248d1e2" ], "author": { "name": "Sebastien Dugue", "email": "sebastien.dugue@bull.net", "time": "Mon Dec 01 14:09:07 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Jan 30 22:35:29 2009 +0100" }, "message": "hrtimers: allow the hot-unplugging of all cpus\n\nImpact: fix CPU hotplug hang on Power6 testbox\n\nOn architectures that support offlining all cpus (at least powerpc/pseries),\nhot-unpluging the tick_do_timer_cpu can result in a system hang.\n\nThis comes from the fact that if the cpu going down happens to be the\ncpu doing the tick, then as the tick_do_timer_cpu handover happens after the\ncpu is dead (via the CPU_DEAD notification), we\u0027re left without ticks,\njiffies are frozen and any task relying on timers (msleep, ...) is stuck.\nThat\u0027s particularly the case for the cpu looping in __cpu_die() waiting\nfor the dying cpu to be dead.\n\nThis patch addresses this by having the tick_do_timer_cpu handover happen\nearlier during the CPU_DYING notification. For this, a new clockevent\nnotification type is introduced (CLOCK_EVT_NOTIFY_CPU_DYING) which is triggered\nin hrtimer_cpu_notify().\n\nSigned-off-by: Sebastien Dugue \u003csebastien.dugue@bull.net\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "6b954823c24f04ed026a8517f6bab5abda279db8", "tree": "bc034346fa51ba63361e23a1631e163eeb52097e", "parents": [ "e7577c50f2fb2d1c167e2c04a4b4c2cc042acb82" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Thu Jan 01 10:12:25 2009 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Thu Jan 01 10:12:25 2009 +1030" }, "message": "cpumask: convert kernel time functions\n\nImpact: Use new APIs\n\nConvert kernel/time functions to use struct cpumask *.\n\nNote the ugly bitmap declarations in tick-broadcast.c. These should\nbe cpumask_var_t, but there was no obvious initialization function to\nput the alloc_cpumask_var() calls in. This was safe.\n\n(Eventually \u0027struct cpumask\u0027 will be undefined for CONFIG_CPUMASK_OFFSTACK,\nso we use a bitmap here to show we really mean it).\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\n" }, { "commit": "320ab2b0b1e08e3805a3e1084a2f0eb1938d5d67", "tree": "1303d8ca53cca655425d512d65cc9ab043254b31", "parents": [ "0de26520c7cabf36e1de090ea8092f011a6106ce" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Sat Dec 13 21:20:26 2008 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Sat Dec 13 21:20:26 2008 +1030" }, "message": "cpumask: convert struct clock_event_device to cpumask pointers.\n\nImpact: change calling convention of existing clock_event APIs\n\nstruct clock_event_timer\u0027s cpumask field gets changed to take pointer,\nas does the -\u003ebroadcast function.\n\nAnother single-patch change. For safety, we BUG_ON() in\nclockevents_register_device() if it\u0027s not set.\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "0de26520c7cabf36e1de090ea8092f011a6106ce", "tree": "3d02e509b6315fdfd9cdb8c9e0b9ed0a30cf9384", "parents": [ "29c0177e6a4ac094302bed54a1d4bbb6b740a9ef" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Sat Dec 13 21:20:26 2008 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Sat Dec 13 21:20:26 2008 +1030" }, "message": "cpumask: make irq_set_affinity() take a const struct cpumask\n\nImpact: change existing irq_chip API\n\nNot much point with gentle transition here: the struct irq_chip\u0027s\nsetaffinity method signature needs to change.\n\nFortunately, not widely used code, but hits a few architectures.\n\nNote: In irq_select_affinity() I save a temporary in by mangling\nirq_desc[irq].affinity directly. Ingo, does this break anything?\n\n(Folded in fix from KOSAKI Motohiro)\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nReviewed-by: Grant Grundler \u003cgrundler@parisc-linux.org\u003e\nAcked-by: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: ralf@linux-mips.org\nCc: grundler@parisc-linux.org\nCc: jeremy@xensource.com\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\n" }, { "commit": "27ce4cb4a0c7cf59b9a9952266883862f2e4c99f", "tree": "066245d42877d8535853e3199b33cdf1110eb8e4", "parents": [ "302745699c1b675b5d2a1af87271de10e4d96b6a" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Mon Sep 22 19:04:02 2008 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Sep 23 11:38:53 2008 +0200" }, "message": "clockevents: prevent mode mismatch on cpu online\n\nImpact: timer hang on CPU online observed on AMD C1E systems\n\nWhen a CPU is brought online then the broadcast machinery can\nbe in the one shot state already. Check this and setup the timer \ndevice of the new CPU in one shot mode so the broadcast code\ncan pick up the next_event value correctly.\n\nAnother AMD C1E oddity, as we switch to broadcast immediately and\nnot after the full bring up via the ACPI cpu idle code.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "6441402b1f173fa38e561d3cee7c01c32e5281ad", "tree": "fd7c81337354f39e44581530d499bcdda06d76c4", "parents": [ "72d31053f62c4bc464c2783974926969614a8649" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Mon Sep 22 18:46:37 2008 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Sep 23 11:38:52 2008 +0200" }, "message": "clockevents: prevent cpu online to interfere with nohz\n\nImpact: rare hang which can be triggered on CPU online.\n\ntick_do_timer_cpu keeps track of the CPU which updates jiffies\nvia do_timer. The value -1 is used to signal, that currently no\nCPU is doing this. There are two cases, where the variable can \nhave this state:\n\n boot:\n necessary for systems where the boot cpu id can be !\u003d 0\n\n nohz long idle sleep:\n When the CPU which did the jiffies update last goes into\n a long idle sleep it drops the update jiffies duty so\n another CPU which is not idle can pick it up and keep\n jiffies going.\n\nUsing the same value for both situations is wrong, as the CPU online\ncode can see the -1 state when the timer of the newly onlined CPU is\nsetup. The setup for a newly onlined CPU goes through periodic mode\nand can pick up the do_timer duty without being aware of the nohz /\nhighres mode of the already running system.\n\nUse two separate states and make them constants to avoid magic\nnumbers confusion. \n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "2344abbcbdb82140050e8be29d3d55e4f6fe860b", "tree": "46c1842fc2a47aa4d7ee0c2c558f54bc50772b69", "parents": [ "f1926ce63b996b42772b39e4b47bb4ef4ba748b4" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Sep 16 11:32:50 2008 -0700" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Sep 16 13:47:02 2008 -0700" }, "message": "clockevents: make device shutdown robust\n\nThe device shut down does not cleanup the next_event variable of the\nclock event device. So when the device is reactivated the possible\nstale next_event value can prevent the device to be reprogrammed as it\nclaims to wait on a event already.\n\nThis is the root cause of the resurfacing suspend/resume problem,\nwhere systems need key press to come back to life.\n\nFix this by setting next_event to KTIME_MAX when the device is shut\ndown. Use a separate function for shutdown which takes care of that\nand only keep the direct set mode call in the broadcast code, where we\ncan not touch the next_event value.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "7c1e76897492d92b6a1c2d6892494d39ded9680c", "tree": "ea04e1be252e176fe5a665a28fd1c26a562cb1fa", "parents": [ "d210baf53b699fc61aa891c177b71d7082d3b957" ], "author": { "name": "Venkatesh Pallipadi", "email": "venkatesh.pallipadi@intel.com", "time": "Wed Sep 03 21:36:50 2008 +0000" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Sep 05 11:11:51 2008 +0200" }, "message": "clockevents: prevent clockevent event_handler ending up handler_noop\n\nThere is a ordering related problem with clockevents code, due to which\nclockevents_register_device() called after tickless/highres switch\nwill not work. The new clockevent ends up with clockevents_handle_noop as\nevent handler, resulting in no timer activity.\n\nThe problematic path seems to be\n\n* old device already has hrtimer_interrupt as the event_handler\n* new clockevent device registers with a higher rating\n* tick_check_new_device() is called\n * clockevents_exchange_device() gets called\n * old-\u003eevent_handler is set to clockevents_handle_noop\n * tick_setup_device() is called for the new device\n * which sets new-\u003eevent_handler using the old-\u003eevent_handler which is noop.\n\nChange the ordering so that new device inherits the proper handler.\n\nThis does not have any issue in normal case as most likely all the clockevent\ndevices are setup before the highres switch. But, can potentially be affecting\nsome corner case where HPET force detect happens after the highres switch.\nThis was a problem with HPET in MSI mode code that we have been experimenting\nwith.\n\nSigned-off-by: Venkatesh Pallipadi \u003cvenkatesh.pallipadi@intel.com\u003e\nSigned-off-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "0bc3cc03fa6e1c20aecb5a33356bcaae410640b9", "tree": "7dab4b0002298b45e31053fe0f37e5ff745682b7", "parents": [ "6524d938b3360504b43a1278b5a8403e85383d1a" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Thu Jul 24 18:21:31 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Jul 26 16:40:33 2008 +0200" }, "message": "cpumask: change cpumask_of_cpu_ptr to use new cpumask_of_cpu\n\n * Replace previous instances of the cpumask_of_cpu_ptr* macros\n with a the new (lvalue capable) generic cpumask_of_cpu().\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Jack Steiner \u003csteiner@sgi.com\u003e\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "c18a41fbbc500ac0307ffd2b0ae73c2af9d0b0ab", "tree": "4242e1bdd8200ef00b5ae5da5d717231fa90a430", "parents": [ "333cdd1f0e5e0aad6b7f8992291563bc7b14670b" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Tue Jul 15 14:14:34 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Jul 18 22:02:59 2008 +0200" }, "message": "cpumask: Optimize cpumask_of_cpu in kernel/time/tick-common.c\n\n * Optimize various places where a pointer to the cpumask_of_cpu value\n will result in reducing stack pressure.\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d7b906897e9caae452947e33674df0a2d6f7e10f", "tree": "5df31094a1b24a46369bda89cdf54d5403b1caa9", "parents": [ "43ca5c3a1cefdaa09231d64485b8f676118bf1e0" ], "author": { "name": "Russell King", "email": "rmk+lkml@arm.linux.org.uk", "time": "Thu Apr 17 07:46:24 2008 +0200" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Thu Apr 17 07:47:05 2008 +0200" }, "message": "[S390] genirq/clockevents: move irq affinity prototypes/inlines to interrupt.h\n\n\u003e Generic code is not supposed to include irq.h. Replace this include\n\u003e by linux/hardirq.h instead and add/replace an include of linux/irq.h\n\u003e in asm header files where necessary.\n\u003e This change should only matter for architectures that make use of\n\u003e GENERIC_CLOCKEVENTS.\n\u003e Architectures in question are mips, x86, arm, sh, powerpc, uml and sparc64.\n\u003e\n\u003e I did some cross compile tests for mips, x86_64, arm, powerpc and sparc64.\n\u003e This patch fixes also build breakages caused by the include replacement in\n\u003e tick-common.h.\n\nI generally dislike adding optional linux/* includes in asm/* includes -\nI\u0027m nervous about this causing include loops.\n\nHowever, there\u0027s a separate point to be discussed here.\n\nThat is, what interfaces are expected of every architecture in the kernel.\nIf generic code wants to be able to set the affinity of interrupts, then\nthat needs to become part of the interfaces listed in linux/interrupt.h\nrather than linux/irq.h.\n\nSo what I suggest is this approach instead (against Linus\u0027 tree of a\ncouple of days ago) - we move irq_set_affinity() and irq_can_set_affinity()\nto linux/interrupt.h, change the linux/irq.h includes to linux/interrupt.h\nand include asm/irq_regs.h where needed (asm/irq_regs.h is supposed to be\nrarely used include since not much touches the stacked parent context\nregisters.)\n\nBuild tested on ARM PXA family kernels and ARM\u0027s Realview platform\nkernels which both use genirq.\n\n[ tglx@linutronix.de: add GENERIC_HARDIRQ dependencies ]\n\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "1595f452f3d8daa066bfd3ba4120754bed3329e1", "tree": "86fa49114c03e0f2dc7c7987b2a5be6adace140f", "parents": [ "b097976e8d6f6e6220161fa6b72b0798ce9f4f4c" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sun Oct 14 22:57:45 2007 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@inhelltoy.tec.linutronix.de", "time": "Sun Oct 14 22:57:45 2007 +0200" }, "message": "clockevents: introduce force broadcast notifier\n\nThe 64bit SMP bootup is slightly different to the 32bit one. It enables\nthe boot CPU local APIC timer before all CPUs are brought up. Some AMD C1E\nsystems have the C1E feature flag only set in the secondary CPU. Due to\nthe early enable of the boot CPU local APIC timer the APIC timer is\nregistered as a fully functional device. When we detect the wreckage during\nthe bringup of the secondary CPU, we need to force the boot CPU into\nbroadcast mode. \n\nAdd a new notifier reason and implement the force broadcast in the clock\nevents layer.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\n" }, { "commit": "4a93232dab0a07074bcc5291a0f1f39919916f31", "tree": "34dbc3209474b2cf5065726dd295c25584f4ce6a", "parents": [ "31c435d75e7d15a5f965c5eb0e33fe0e236f49a4" ], "author": { "name": "Venki Pallipadi", "email": "venkatesh.pallipadi@intel.com", "time": "Fri Oct 12 23:04:23 2007 +0200" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@inhelltoy.tec.linutronix.de", "time": "Fri Oct 12 23:04:23 2007 +0200" }, "message": "clock events: allow replacement of broadcast timer\n\nChange the broadcast timer, if a timer with higher rating becomes available.\n\nSigned-off-by: Venkatesh Pallipadi \u003cvenkatesh.pallipadi@intel.com\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Greg KH \u003cgreg@kroah.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\n" }, { "commit": "18de5bc4c1f1f1fa5e14f354a7603bd6e9d4e3b6", "tree": "cb0ff399fb3f368adecf213318e32a499ada6ff9", "parents": [ "93da56efcf8c6a111f0349f6b7651172d4745ca0" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sat Jul 21 04:37:34 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sat Jul 21 17:49:15 2007 -0700" }, "message": "clockevents: fix resume logic\n\nWe need to make sure, that the clockevent devices are resumed, before\nthe tick is resumed. The current resume logic does not guarantee this.\n\nAdd CLOCK_EVT_MODE_RESUME and call the set mode functions of the clock\nevent devices before resuming the tick / oneshot functionality.\n\nFixup the existing users.\n\nThanks to Nigel Cunningham for tracking down a long standing thinko,\nwhich affected the jinxed VAIO.\n\n[akpm@linux-foundation.org: xen build fix]\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d3ed782458f315c30ea679b919a2cc59f2b82565", "tree": "b87fffc87acf5632566a6384f5c8be8f5c2e03b2", "parents": [ "d5d3b736e3264934ec832a657a9a434b65f3d51f" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue May 08 00:30:03 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue May 08 11:15:10 2007 -0700" }, "message": "highres/dyntick: prevent xtime lock contention\n\nWhile the !highres/!dyntick code assigns the duty of the do_timer() call to\none specific CPU, this was dropped in the highres/dyntick part during\ndevelopment.\n\nSteven Rostedt discovered the xtime lock contention on highres/dyntick due\nto several CPUs trying to update jiffies.\n\nAdd the single CPU assignement back. In the dyntick case this needs to be\nhandled carefully, as the CPU which has the do_timer() duty must drop the\nassignement and let it be grabbed by another CPU, which is active.\nOtherwise the do_timer() calls would not happen during the long sleep.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nAcked-by: Mark Lord \u003cmlord@pobox.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "cd05a1f818073a623455a58e756c5b419fc98db9", "tree": "4e148e96f00fe07b0c53a379e812344733e8484a", "parents": [ "24c4ac070adffe4a21f3a8daf4aee7c98fa6c4f9" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sat Mar 17 00:25:52 2007 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Mar 16 19:35:25 2007 -0700" }, "message": "[PATCH] clockevents: Fix suspend/resume to disk hangs\n\nI finally found a dual core box, which survives suspend/resume without\ncrashing in the middle of nowhere. Sigh, I never figured out from the\ncode and the bug reports what\u0027s going on.\n\nThe observed hangs are caused by a stale state transition of the clock\nevent devices, which keeps the RCU synchronization away from completion,\nwhen the non boot CPU is brought back up.\n\nThe suspend/resume in oneshot mode needs the similar care as the\nperiodic mode during suspend to RAM. My assumption that the state\ntransitions during the different shutdown/bringups of s2disk would go\nthrough the periodic boot phase and then switch over to highres resp.\nnohz mode were simply wrong.\n\nAdd the appropriate suspend / resume handling for the non periodic\nmodes.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6321dd60c76b2e12383bc06046288b15397ed3a0", "tree": "4cea63e69ce4f6f5f7f94fc6905063efdde7974e", "parents": [ "c3442e296517aee733d62fc3fe03211598902c7d" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Tue Mar 06 08:25:42 2007 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Mar 06 09:30:24 2007 -0800" }, "message": "[PATCH] Save/restore periodic tick information over suspend/resume\n\nThe programming of periodic tick devices needs to be saved/restored\nacross suspend/resume - otherwise we might end up with a system coming\nup that relies on getting a PIT (or HPET) interrupt, while those devices\ndefault to \u0027no interrupts\u0027 after powerup. (To confuse things it worked\nto a certain degree on some systems because the lapic gets initialized\nas a side-effect of SMP bootup.)\n\nThis suspend / resume thing was dropped unintentionally during the\nlast-minute -mm code reshuffling.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3494c16676a21e7e53e21b08a0a469a38df6dcfb", "tree": "d80452cfe2d109e10910afcf735b68d5de6ec67b", "parents": [ "9e203bcc1051cac2a8b15c3ee9db4c0d05794abe" ], "author": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Feb 24 22:11:42 2007 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Feb 26 11:14:15 2007 -0800" }, "message": "[TICK] tick-common: Fix one-shot handling in tick_handle_periodic().\n\nWhen clockevents_program_event() is given an expire time in the\npast, it does not update dev-\u003enext_event, so this looping code\nwould loop forever once the first in-the-past expiration time\nwas used.\n\nKeep advancing \"next\" locally to fix this bug.\n\nAcked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "289f480af87e45f7a6de6ba9b4c061c2e259fe98", "tree": "1075cdb8656d7af12558ea2a3dee1fcd1d7a44a7", "parents": [ "82f67cd9fca8c8762c15ba7ed0d5747588c1e221" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Feb 16 01:28:15 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 16 08:13:59 2007 -0800" }, "message": "[PATCH] Add debugging feature /proc/timer_list\n\nadd /proc/timer_list, which prints all currently pending (high-res) timers,\nall clock-event sources and their parameters in a human-readable form.\n\nSample output:\n\nTimer List Version: v0.1\nHRTIMER_MAX_CLOCK_BASES: 2\nnow at 4246046273872 nsecs\n\ncpu: 0\n clock 0:\n .index: 0\n .resolution: 1 nsecs\n .get_time: ktime_get_real\n .offset: 1273998312645738432 nsecs\nactive timers:\n clock 1:\n .index: 1\n .resolution: 1 nsecs\n .get_time: ktime_get\n .offset: 0 nsecs\nactive timers:\n #0: \u003cf5a90ec8\u003e, hrtimer_sched_tick, hrtimer_stop_sched_tick, swapper/0\n # expires at 4246432689566 nsecs [in 386415694 nsecs]\n #1: \u003cf5a90ec8\u003e, hrtimer_wakeup, do_nanosleep, pcscd/2050\n # expires at 4247018194689 nsecs [in 971920817 nsecs]\n #2: \u003cf5a90ec8\u003e, hrtimer_wakeup, do_nanosleep, irqbalance/1909\n # expires at 4247351358392 nsecs [in 1305084520 nsecs]\n #3: \u003cf5a90ec8\u003e, hrtimer_wakeup, do_nanosleep, crond/2157\n # expires at 4249097614968 nsecs [in 3051341096 nsecs]\n #4: \u003cf5a90ec8\u003e, it_real_fn, do_setitimer, syslogd/1888\n # expires at 4251329900926 nsecs [in 5283627054 nsecs]\n .expires_next : 4246432689566 nsecs\n .hres_active : 1\n .check_clocks : 0\n .nr_events : 31306\n .idle_tick : 4246020791890 nsecs\n .tick_stopped : 1\n .idle_jiffies : 986504\n .idle_calls : 40700\n .idle_sleeps : 36014\n .idle_entrytime : 4246019418883 nsecs\n .idle_sleeptime : 4178181972709 nsecs\n\ncpu: 1\n clock 0:\n .index: 0\n .resolution: 1 nsecs\n .get_time: ktime_get_real\n .offset: 1273998312645738432 nsecs\nactive timers:\n clock 1:\n .index: 1\n .resolution: 1 nsecs\n .get_time: ktime_get\n .offset: 0 nsecs\nactive timers:\n #0: \u003cf5a90ec8\u003e, hrtimer_sched_tick, hrtimer_restart_sched_tick, swapper/0\n # expires at 4246050084568 nsecs [in 3810696 nsecs]\n #1: \u003cf5a90ec8\u003e, hrtimer_wakeup, do_nanosleep, atd/2227\n # expires at 4261010635003 nsecs [in 14964361131 nsecs]\n #2: \u003cf5a90ec8\u003e, hrtimer_wakeup, do_nanosleep, smartd/2332\n # expires at 5469485798970 nsecs [in 1223439525098 nsecs]\n .expires_next : 4246050084568 nsecs\n .hres_active : 1\n .check_clocks : 0\n .nr_events : 24043\n .idle_tick : 4246046084568 nsecs\n .tick_stopped : 0\n .idle_jiffies : 986510\n .idle_calls : 26360\n .idle_sleeps : 22551\n .idle_entrytime : 4246043874339 nsecs\n .idle_sleeptime : 4170763761184 nsecs\n\ntick_broadcast_mask: 00000003\nevent_broadcast_mask: 00000001\n\nCPU#0\u0027s local event device:\n\nClock Event Device: lapic\n capabilities: 0000000e\n max_delta_ns: 807385544\n min_delta_ns: 1443\n mult: 44624025\n shift: 32\n set_next_event: lapic_next_event\n set_mode: lapic_timer_setup\n event_handler: hrtimer_interrupt\n .installed: 1\n .expires: 4246432689566 nsecs\n\nCPU#1\u0027s local event device:\n\nClock Event Device: lapic\n capabilities: 0000000e\n max_delta_ns: 807385544\n min_delta_ns: 1443\n mult: 44624025\n shift: 32\n set_next_event: lapic_next_event\n set_mode: lapic_timer_setup\n event_handler: hrtimer_interrupt\n .installed: 1\n .expires: 4246050084568 nsecs\n\nClock Event Device: hpet\n capabilities: 00000007\n max_delta_ns: 2147483647\n min_delta_ns: 3352\n mult: 61496110\n shift: 32\n set_next_event: hpet_next_event\n set_mode: hpet_set_mode\n event_handler: handle_nextevt_broadcast\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Roman Zippel \u003czippel@linux-m68k.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "79bf2bb335b85db25d27421c798595a2fa2a0e82", "tree": "550ec2654ae1dd65b871de7fe9c890108c6e86d8", "parents": [ "f8381cba04ba8173fd5a2b8e5cd8b3290ee13a98" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Feb 16 01:28:03 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 16 08:13:59 2007 -0800" }, "message": "[PATCH] tick-management: dyntick / highres functionality\n\nWith Ingo Molnar \u003cmingo@elte.hu\u003e\n\nAdd functions to provide dynamic ticks and high resolution timers. The code\nwhich keeps track of jiffies and handles the long idle periods is shared\nbetween tick based and high resolution timer based dynticks. The dyntick\nfunctionality can be disabled on the kernel commandline. Provide also the\ninfrastructure to support high resolution timers.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Roman Zippel \u003czippel@linux-m68k.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f8381cba04ba8173fd5a2b8e5cd8b3290ee13a98", "tree": "ad8c9f91ce031a04c62ff75fcd3237fc666f1c2e", "parents": [ "906568c9c668ff994f4078932ec6ae1e3950d1af" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Feb 16 01:28:02 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 16 08:13:59 2007 -0800" }, "message": "[PATCH] tick-management: broadcast functionality\n\nWith Ingo Molnar \u003cmingo@elte.hu\u003e\n\nAdd broadcast functionality, so per cpu clock event devices can be registered\nas dummy devices or switched from/to broadcast on demand. The broadcast\nfunction distributes the events via the broadcast function of the clock event\ndevice. This is primarily designed to replace the switch apic timer to / from\nIPI in power states, where the apic stops.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Roman Zippel \u003czippel@linux-m68k.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "906568c9c668ff994f4078932ec6ae1e3950d1af", "tree": "825ac33fc61af55a0fe7485a1df681f5a6126d7b", "parents": [ "d316c57ff6bfad9557462b9100f25c6260d2b774" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Feb 16 01:28:01 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 16 08:13:59 2007 -0800" }, "message": "[PATCH] tick-management: core functionality\n\nWith Ingo Molnar \u003cmingo@elte.hu\u003e\n\nThe tick-management code is the first user of the clockevents layer. It takes\nclock event devices from the clock events core and uses them to provide the\nperiodic tick.\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: john stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Roman Zippel \u003czippel@linux-m68k.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ] }