)]}' { "log": [ { "commit": "7a232e0350940d2664f4de5cc3f0f443bae5062d", "tree": "bb91a53faab7bf6e92df127f8bdb2225cf808925", "parents": [ "2e084786f6fe052274f1dfa7c675fe4a02cacd6e" ], "author": { "name": "Lai Jiangshan", "email": "laijs@cn.fujitsu.com", "time": "Thu Jun 12 16:43:07 2008 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Jun 12 14:29:54 2008 +0200" }, "message": "sched: 64-bit: fix arithmetics overflow\n\n(overflow means weight \u003e\u003d 2^32 here, because inv_weigh \u003d 2^32/weight)\n\nA weight of a cfs_rq is the sum of weights of which entities\nare queued on this cfs_rq, so it will overflow when there are\ntoo many entities.\n\nAlthough, overflow occurs very rarely, but it break fairness when\nit occurs. 64-bits systems have more memory than 32-bit systems\nand 64-bit systems can create more process usually, so overflow may\noccur more frequently.\n\nThis patch guarantees fairness when overflow happens on 64-bit systems.\nThanks to the optimization of compiler, it changes nothing on 32-bit.\n\nSigned-off-by: Lai Jiangshan \u003claijs@cn.fujitsu.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "2e084786f6fe052274f1dfa7c675fe4a02cacd6e", "tree": "66190825236b1f3f59e88aaa83d2683bd121143d", "parents": [ "16882c1e962b4be5122fc05aaf2afc10fd9e2d15" ], "author": { "name": "Lai Jiangshan", "email": "laijs@cn.fujitsu.com", "time": "Thu Jun 12 16:42:58 2008 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Jun 12 14:23:55 2008 +0200" }, "message": "sched: fair group: fix overflow(was: fix divide by zero)\n\nI found a bug which can be reproduced by this way:(linux-2.6.26-rc5, x86-64)\n(use 2^32, 2^33, ...., 2^63 as shares value)\n\n# mkdir /dev/cpuctl\n# mount -t cgroup -o cpu cpuctl /dev/cpuctl\n# cd /dev/cpuctl\n# mkdir sub\n# echo 0x8000000000000000 \u003e sub/cpu.shares\n# echo $$ \u003e sub/tasks\noops here! divide by zero.\n\nThis is because do_div() expects the 2th parameter to be 32 bits,\nbut unsigned long is 64 bits in x86_64.\n\nPeter Zijstra pointed it out that the sane thing to do is limit the\nshares value to something smaller instead of using an even more\nexpensive divide.\n\nAlso, I found another bug about \"the shares value is too large\":\n\npid1 and pid2 are set affinity to cpu#0\npid1 is attached to cg1 and pid2 is attached to cg2\n\nif cg1/cpu.shares \u003d 1024 cg2/cpu.shares \u003d 2000000000\nthen pid2 got 100% usage of cpu, and pid1 0%\n\nif cg1/cpu.shares \u003d 1024 cg2/cpu.shares \u003d 20000000000\nthen pid2 got 0% usage of cpu, and pid1 100%\n\nAnd a weight of a cfs_rq is the sum of weights of which entities\nare queued on this cfs_rq, so the shares value should be limited\nto a smaller value.\n\nI think that (1UL \u003c\u003c 18) is a good limited value:\n\n1) it\u0027s not too large, we can create a lot of group before overflow\n2) it\u0027s several times the weight value for nice\u003d-19 (not too small)\n\nSigned-off-by: Lai Jiangshan \u003claijs@cn.fujitsu.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "16882c1e962b4be5122fc05aaf2afc10fd9e2d15", "tree": "c39cae4ae4874998d7e3486cd57d57613b05a89c", "parents": [ "39b945a37bac2b692773a470890c8ba301485b15" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Sun Jun 08 21:20:41 2008 +0400" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Jun 10 11:37:25 2008 +0200" }, "message": "sched: fix TASK_WAKEKILL vs SIGKILL race\n\nschedule() has the special \"TASK_INTERRUPTIBLE \u0026\u0026 signal_pending()\" case,\nthis allows us to do\n\n\tcurrent-\u003estate \u003d TASK_INTERRUPTIBLE;\n\tschedule();\n\nwithout fear to sleep with pending signal.\n\nHowever, the code like\n\n\tcurrent-\u003estate \u003d TASK_KILLABLE;\n\tschedule();\n\nis not right, schedule() doesn\u0027t take TASK_WAKEKILL into account. This means\nthat mutex_lock_killable(), wait_for_completion_killable(), down_killable(),\nschedule_timeout_killable() can miss SIGKILL (and btw the second SIGKILL has\nno effect).\n\nIntroduce the new helper, signal_pending_state(), and change schedule() to\nuse it. Hopefully it will have more users, that is why the task\u0027s state is\npassed separately.\n\nNote this \"__TASK_STOPPED | __TASK_TRACED\" check in signal_pending_state().\nThis is needed to preserve the current behaviour (ptrace_notify). I hope\nthis check will be removed soon, but this (afaics good) change needs the\nseparate discussion.\n\nThe fast path is \"(state \u0026 (INTERRUPTIBLE | WAKEKILL)) + signal_pending(p)\",\nbasically the same that schedule() does now. However, this patch of course\nbloats schedule().\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "6363ca57c76b7b83639ca8c83fc285fa26a7880e", "tree": "b8630b4af286409efdd648920a546fae24d4db88", "parents": [ "4285f594f84d1f0641fc962d00e6638dec4a19c4" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:28:57 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:28:57 2008 +0200" }, "message": "revert (\"sched: fair-group: SMP-nice for group scheduling\")\n\nYanmin Zhang reported:\n\nComparing with 2.6.25, volanoMark has big regression with kernel 2.6.26-rc1.\nIt\u0027s about 50% on my 8-core stoakley, 16-core tigerton, and Itanium Montecito.\n\nWith bisect, I located the following patch:\n\n| 18d95a2832c1392a2d63227a7a6d433cb9f2037e is first bad commit\n| commit 18d95a2832c1392a2d63227a7a6d433cb9f2037e\n| Author: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n| Date: Sat Apr 19 19:45:00 2008 +0200\n|\n| sched: fair-group: SMP-nice for group scheduling\n\nRevert it so that we get v2.6.25 behavior.\n\nBisected-by: Yanmin Zhang \u003cyanmin_zhang@linux.intel.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4285f594f84d1f0641fc962d00e6638dec4a19c4", "tree": "cb00af70f896e89bda957ddbf5123fe551d96698", "parents": [ "c6fba5451a84143f34056a465e72ba187fcc651c" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri May 16 17:47:14 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:25:15 2008 +0200" }, "message": "sched: cleanup\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "3f33a7ce9567ded582af1ab71f9802165fe12f09", "tree": "04304757dc84bc36400f306677d3452635cf9f74", "parents": [ "f9305d4a0968201b2818dbed0dc8cb0d4ee7aeb3" ], "author": { "name": "Roel Kluin", "email": "roel.kluin@gmail.com", "time": "Tue May 13 23:44:11 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:25:14 2008 +0200" }, "message": "sched: unite unlikely pairs in rt_policy() and schedule_debug()\n\nRemoves obfuscation and may improve assembly.\n\nSigned-off-by: Roel Kluin \u003croel.kluin@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f9305d4a0968201b2818dbed0dc8cb0d4ee7aeb3", "tree": "e171e568f04bd25c7c2ff99b5ef673b917b6eae6", "parents": [ "f26a3988917913b3d11b2bd741601a2c64ab9204" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:23:17 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu May 29 11:24:01 2008 +0200" }, "message": "revert (\"sched: fair: weight calculations\")\n\nYanmin Zhang reported:\n\nComparing with kernel 2.6.25, sysbench+mysql(oltp, readonly) has many\nregressions with 2.6.26-rc1:\n\n 1) 8-core stoakley: 28%;\n 2) 16-core tigerton: 20%;\n 3) Itanium Montvale: 50%.\n\nBisect located this patch:\n\n| 8f1bc385cfbab474db6c27b5af1e439614f3025c is first bad commit\n| commit 8f1bc385cfbab474db6c27b5af1e439614f3025c\n| Author: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n| Date: Sat Apr 19 19:45:00 2008 +0200\n|\n| sched: fair: weight calculations\n\nRevert it to the 2.6.25 state.\n\nBisected-by: Yanmin Zhang \u003cyanmin_zhang@linux.intel.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "0c70814c311581a6c86198db4f982aa683c68fb8", "tree": "3df20cc52e4cf0515b7d7f6658d9f6991baa560c", "parents": [ "b7cffc1f29c1bc729bc50c863c87f93f9b70994b" ], "author": { "name": "Mirco Tischler", "email": "mt-ml@gmx.de", "time": "Wed May 14 16:05:46 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 14 19:11:14 2008 -0700" }, "message": "cgroups: fix compile warning\n\nReturn type of cpu_rt_runtime_write() should be int instead of ssize_t.\n\nSigned-off-by: Mirco Tischler \u003cmt-ml@gmx.de\u003e\nAcked-by: Paul Menage \u003cmenage@google.com\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": "c3921ab71507b108d51a0f1ee960f80cd668a93d", "tree": "b1408b898a8b50f15ad4a0cf1f29e17cc0138485", "parents": [ "9662369786b9d07fd46d65b0f9e3938a3e01a5d9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 11 16:04:48 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 11 16:04:48 2008 -0700" }, "message": "Add new \u0027cond_resched_bkl()\u0027 helper function\n\nIt acts exactly like a regular \u0027cond_resched()\u0027, but will not get\noptimized away when CONFIG_PREEMPT is set.\n\nNormal kernel code is already preemptable in the presense of\nCONFIG_PREEMPT, so cond_resched() is optimized away (see commit\n02b67cc3ba36bdba351d6c3a00593f4ec550d9d3 \"sched: do not do\ncond_resched() when CONFIG_PREEMPT\").\n\nBut when wanting to conditionally reschedule while holding a lock, you\nneed to use \"cond_sched_lock(lock)\", and the new function is the BKL\nequivalent of that.\n\nAlso make fs/locks.c use it.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8e3e076c5a78519a9f64cd384e8f18bc21882ce0", "tree": "f032258fde3aa4771e86bf4552fe4530c221dec3", "parents": [ "00b41ec2611dc98f87f30753ee00a53db648d662" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 10 20:58:02 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 10 20:58:02 2008 -0700" }, "message": "BKL: revert back to the old spinlock implementation\n\nThe generic semaphore rewrite had a huge performance regression on AIM7\n(and potentially other BKL-heavy benchmarks) because the generic\nsemaphores had been rewritten to be simple to understand and fair. The\nlatter, in particular, turns a semaphore-based BKL implementation into a\nmess of scheduling.\n\nThe attempt to fix the performance regression failed miserably (see the\nprevious commit 00b41ec2611dc98f87f30753ee00a53db648d662 \u0027Revert\n\"semaphore: fix\"\u0027), and so for now the simple and sane approach is to\ninstead just go back to the old spinlock-based BKL implementation that\nnever had any issues like this.\n\nThis patch also has the advantage of being reported to fix the\nregression completely according to Yanmin Zhang, unlike the semaphore\nhack which still left a couple percentage point regression.\n\nAs a spinlock, the BKL obviously has the potential to be a latency\nissue, but it\u0027s not really any different from any other spinlock in that\nrespect. We do want to get rid of the BKL asap, but that has been the\nplan for several years.\n\nThese days, the biggest users are in the tty layer (open/release in\nparticular) and Alan holds out some hope:\n\n \"tty release is probably a few months away from getting cured - I\u0027m\n afraid it will almost certainly be the very last user of the BKL in\n tty to get fixed as it depends on everything else being sanely locked.\"\n\nso while we\u0027re not there yet, we do have a plan of action.\n\nTested-by: Yanmin Zhang \u003cyanmin_zhang@linux.intel.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Matthew Wilcox \u003cmatthew@wil.cx\u003e\nCc: Alexander Viro \u003cviro@ftp.linux.org.uk\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3e51f33fcc7f55e6df25d15b55ed10c8b4da84cd", "tree": "3752f9ea8e014ec40e95a1b197b0a3d18e1056a8", "parents": [ "a5574cf65b5f03ce9ade3918764fe22e5e2371e3" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat May 03 18:29:28 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: add optional support for CONFIG_HAVE_UNSTABLE_SCHED_CLOCK\n\nthis replaces the rq-\u003eclock stuff (and possibly cpu_clock()).\n\n - architectures that have an \u0027imperfect\u0027 hardware clock can set\n CONFIG_HAVE_UNSTABLE_SCHED_CLOCK\n\n - the \u0027jiffie\u0027 window might be superfulous when we update tick_gtod\n before the __update_sched_clock() call in sched_clock_tick()\n\n - cpu_clock() might be implemented as:\n\n sched_clock_cpu(smp_processor_id())\n\n if the accuracy proves good enough - how far can TSC drift in a\n single jiffie when considering the filtering and idle hooks?\n\n[ mingo@elte.hu: various fixes and cleanups ]\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "dfbf4a1bc319f0f9a31e39b2da1fa5c55e85af89", "tree": "0b9dd19406c53a93452dd345bb05f76aa712a757", "parents": [ "cb4ad1ffc7c0d8ea7dc8cd8ba303d83551716d46" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Apr 23 09:24:06 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: fix cpu clock\n\nDavid Miller pointed it out that nothing in cpu_clock() sets\nprev_cpu_time. This caused __sync_cpu_clock() to be called\nall the time - against the intention of this code.\n\nThe result was that in practice we hit a global spinlock every\ntime cpu_clock() is called - which - even though cpu_clock()\nis used for tracing and debugging, is suboptimal.\n\nWhile at it, also:\n\n- move the irq disabling to the outest layer,\n this should make cpu_clock() warp-free when called with irqs\n enabled.\n\n- use long long instead of cycles_t - for platforms where cycles_t\n is 32-bit.\n\nReported-by: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "cb4ad1ffc7c0d8ea7dc8cd8ba303d83551716d46", "tree": "79f6b1fe971a270e5e009d695d2bf998936197c8", "parents": [ "712555ee4f873515612f89554ad1a3fda5fa887e" ], "author": { "name": "Miao Xie", "email": "miaox@cn.fujitsu.com", "time": "Mon Apr 28 12:54:56 2008 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: fair-group: fix a Div0 error of the fair group scheduler\n\nWhen I echoed 0 into the \"cpu.shares\" file, a Div0 error occured.\n\nWe found it is caused by the following calling.\n\n sched_group_set_shares(tg, shares)\n set_se_shares(tg-\u003ese[i], shares/nr_cpu_ids)\n __set_se_shares(se, shares)\n div64_64((1ULL\u003c\u003c32), shares)\n\nWhen the echoed value was less than the number of processores, the result of the\nsentence \"shares/nr_cpu_ids\" was 0, and then the system called div64() to divide\nthe result, the Div0 error occured.\n\nIt is unnecessary that the shares value is divided by nr_cpu_ids, I think.\nBecause in the function __update_group_shares_cpu() and init_tg_cfs_entry(),\nthe shares value isn\u0027t divided by nr_cpu_ids when setting shares of the sched\nentity.\n\nThis patch fixes this bug. And echoing ULONG_MAX value into cpu.shares also\ncauses Div0 error, so we set a macro MAX_SHARES to limit the max value of\nshares.\n\nSigned-off-by: Miao Xie \u003cmiaox@cn.fujitsu.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "712555ee4f873515612f89554ad1a3fda5fa887e", "tree": "88ff7222727c63959eb58a3aa729029aaf08a64f", "parents": [ "690229a0912ca2fef8b542fe4d8b73acfcdc6e24" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Mon Apr 28 11:33:07 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: fix missing locking in sched_domains code\n\nConcurrent calls to detach_destroy_domains and arch_init_sched_domains\nwere prevented by the old scheduler subsystem cpu hotplug mutex. When\nthis got converted to get_online_cpus() the locking got broken.\nUnlike before now several processes can concurrently enter the critical\nsections that were protected by the old lock.\n\nSo use the already present doms_cur_mutex to protect these sections again.\n\nCc: Gautham R Shenoy \u003cego@in.ibm.com\u003e\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "690229a0912ca2fef8b542fe4d8b73acfcdc6e24", "tree": "28be1db230f039589650f2b5d46de9e6f5722ef0", "parents": [ "d7dcdc11cfa6a8860a29b09f985467b89224699d" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Apr 23 09:31:35 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: make clock sync tunable by architecture code\n\nmake time_sync_thresh tunable to architecture code.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "673a90a1e05c8127886f7659d1a457169378371f", "tree": "77469d2b1273d960b90520a088bd82bdf0276ee7", "parents": [ "b328ca182f01c2a04b85e0ee8a410720b104fbcc" ], "author": { "name": "David Simner", "email": "djs203@srcf.ucam.org", "time": "Tue Apr 29 10:08:59 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: fix sched_info_switch not being called according to documentation\n\nhttp://bugzilla.kernel.org/show_bug.cgi?id\u003d10545\n\nsched_stats.h says that __sched_info_switch is \"called when prev !\u003d\nnext\" in the comment. sched.c should therefore do that.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "b328ca182f01c2a04b85e0ee8a410720b104fbcc", "tree": "d04ba3472a8f57788fb7e6e614e8e6b3c5a79b55", "parents": [ "104f64549c961a797ff5f7c59946a7caa335c5b0" ], "author": { "name": "Peter Zijlstra", "email": "peterz@infradead.org", "time": "Tue Apr 29 10:02:46 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:18 2008 +0200" }, "message": "sched: fix hrtick_start_fair and CPU-Hotplug\n\nGautham R Shenoy reported:\n\n \u003e While running the usual CPU-Hotplug stress tests on linux-2.6.25,\n \u003e I noticed the following in the console logs.\n \u003e\n \u003e This is a wee bit difficult to reproduce. In the past 10 runs I hit this\n \u003e only once.\n \u003e\n \u003e ------------[ cut here ]------------\n \u003e\n \u003e WARNING: at kernel/sched.c:962 hrtick+0x2e/0x65()\n \u003e\n \u003e Just wondering if we are doing a good job at handling the cancellation\n \u003e of any per-cpu scheduler timers during CPU-Hotplug.\n\nThis looks like its indeed not cancelled at all and migrates the it to\nanother cpu. Fix it via a proper hotplug notifier mechanism.\n\nReported-by: Gautham R Shenoy \u003cego@in.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: stable@kernel.org\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "983ed7a66bcec9dc307d89dc7af47cdf209e56af", "tree": "a5fc14c7fe993b795198bbeee2919ad695735e81", "parents": [ "d478c2cfaa2476f8b6876f9eb4d8fddcfa986479" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Thu Apr 24 18:17:55 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:17 2008 +0200" }, "message": "sched: add statics, don\u0027t return void expressions\n\nNoticed by sparse:\nkernel/sched.c:760:20: warning: symbol \u0027sched_feat_names\u0027 was not declared. Should it be static?\nkernel/sched.c:767:5: warning: symbol \u0027sched_feat_open\u0027 was not declared. Should it be static?\nkernel/sched_fair.c:845:3: warning: returning void-valued expression\nkernel/sched.c:4386:3: warning: returning void-valued expression\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d478c2cfaa2476f8b6876f9eb4d8fddcfa986479", "tree": "59cb264c539147cb481e13335ff3c69750e16c67", "parents": [ "733a0771df46af942b8355cd8bb15780106b4353" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Sat Apr 26 11:30:34 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:17 2008 +0200" }, "message": "sched: add debug checks to idle functions\n\nCc: Venkatesh Pallipadi \u003cvenkatesh.pallipadi@intel.com\u003e\nCc: \"Justin Mattock\" \u003cjustinmattock@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "e05510d01ad1565e5e086a939261084d67ba2b10", "tree": "f6a8d27f2ad546c98079c8868d914de46a77994f", "parents": [ "a992241de614dd2b7c97a9ba64e28c0e563f19bf" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Mon May 05 23:56:17 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 05 23:56:17 2008 +0200" }, "message": "sched: optimize calc_delta_mine()\n\nJoel noticed that the !lw-\u003einv_weight contition isn\u0027t unlikely anymore so\nremove the unlikely annotation. Also, remove the two div64_u64() inv_weight\ncalculations, which makes them rely on the calc_delta_mine() path as well.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCC: Joel Schopp \u003cjschopp@austin.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "6f6d6a1a6a1336431a6cba60ace9e97c3a496a19", "tree": "f32e82fc3a50b6877afa3220bdb6f7ea0582e07f", "parents": [ "71abb3af62dfa52930755f3b6497eafbe1d6ec85" ], "author": { "name": "Roman Zippel", "email": "zippel@linux-m68k.org", "time": "Thu May 01 04:34:28 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 01 08:03:58 2008 -0700" }, "message": "rename div64_64 to div64_u64\n\nRename div64_64 to div64_u64 to make it consistent with the other divide\nfunctions, so it clearly includes the type of the divide. Move its definition\nto math64.h as currently no architecture overrides the generic implementation.\n They can still override it of course, but the duplicated declarations are\navoided.\n\nSigned-off-by: Roman Zippel \u003czippel@linux-m68k.org\u003e\nCc: Avi Kivity \u003cavi@qumranet.com\u003e\nCc: Russell King \u003crmk@arm.linux.org.uk\u003e\nCc: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Jeff Dike \u003cjdike@addtoit.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "06ecb27cfbf53ac2c7e397aa1619a6f9a98c5896", "tree": "1633e77714da0d6478b79bd74d95c7fc715db15c", "parents": [ "e73d2c61d1fcbd3621688ae457b49509c8d4c601" ], "author": { "name": "Paul Menage", "email": "menage@google.com", "time": "Tue Apr 29 01:00:06 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 29 08:06:09 2008 -0700" }, "message": "CGroups _s64 files: use read_s64/write_s64 in CFS cgroup for rt_runtime file\n\nThis removes some filesystem boilerplate from the CFS cgroup subsystem.\n\nSigned-off-by: Paul Menage \u003cmenage@google.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\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": "f4c753b7eacc277e506066abdda351cbc1cf8e6a", "tree": "2510264d4fac836dc64acd700846a5fe8d960b1e", "parents": [ "3ff31d0cca38b3c20e88a022bf38c4f7c98492f0" ], "author": { "name": "Paul Menage", "email": "menage@google.com", "time": "Tue Apr 29 00:59:56 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 29 08:06:07 2008 -0700" }, "message": "CGroup API files: rename read/write_uint methods to read_write_u64\n\nSeveral people have justifiably complained that the \"_uint\" suffix is\ninappropriate for functions that handle u64 values, so this patch just renames\nall these functions and their users to have the suffic _u64.\n\n[peterz@infradead.org: build fix]\nSigned-off-by: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Li Zefan\" \u003clizf@cn.fujitsu.com\u003e\nCc: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: \"YAMAMOTO Takashi\" \u003cyamamoto@valinux.co.jp\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6e18933f2b6156d0a0ec9d5522ab6a6033cf7241", "tree": "f8045942c4b5c250fd4eec8331fec8a7e2962974", "parents": [ "eb855fd8e7f6b1d8b50f24aa36803e29b367b041", "03970f065d9b4b156d0e879f82989440f7045396" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 25 12:50:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 25 12:50:00 2008 -0700" }, "message": "Merge branch \u0027release\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6\n\n* \u0027release\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:\n [PATCH] Build fix for CONFIG_NUMA\u003dy \u0026\u0026 CONFIG_SMP\u003dn\n [IA64] fix bootmem regression on Altix\n" }, { "commit": "0b79dada976198cb1a4c043068e3b44d5cab2a5a", "tree": "fde938c698545a7c8764a2eac4304dee002bb281", "parents": [ "50be4917ee70218f59e04dec029121b97fb9cb3d", "3f5087a2bae5d1ce10a3d698dec8f879a96f5419" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 25 12:47:56 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 25 12:47:56 2008 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched-fixes\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched-fixes:\n sched: fix share (re)distribution\n softlockup: fix NOHZ wakeup\n seqlock: livelock fix\n" }, { "commit": "5a9d3225a0d7060bdf3a18018992dc8cef958425", "tree": "585102b8d74e5b1631eeaec89ba428a4019d1697", "parents": [ "b69d3987f4360a5e7e9e55465b3cdd0cc204b79e" ], "author": { "name": "David Miller", "email": "davem@davemloft.net", "time": "Thu Apr 24 20:46:20 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Apr 25 09:53:06 2008 +0200" }, "message": "sched: use alloc_bootmem() instead of alloc_bootmem_low()\n\nThere is no guarantee that there is physical ram below 4GB, and in\nfact many boxes don\u0027t have exactly that.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "3f5087a2bae5d1ce10a3d698dec8f879a96f5419", "tree": "ad28e2dd5d36e7ea435032dd8a5fbd94340342ca", "parents": [ "126e01bf92dfc5f0ba91e88be02c473e1506d7d9" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Apr 25 00:25:08 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Apr 25 00:25:08 2008 +0200" }, "message": "sched: fix share (re)distribution\n\nfix __aggregate_redistribute_shares() related lockup reported by\nDavid S. Miller.\n\nThe problem this code tries to solve is \u0027accurately\u0027 calculating the \u0027fair\u0027\nshare of the group weight for each cpu. The current code falls back to a global\ngroup rebalance in case the sched_domain\u0027s span it looks at has no shares, but\ndoes have tasks.\n\nThe reason it gets stuck here, is because its inherently racy - if someone\nsteals the last task after we compute the agg-\u003erq_weight, but before we\nrebalance, we\u0027ll never get out of the loop.\n\nWe could of course go fix that, but while looking at this issue I found that\nthis \u0027fallback\u0027 wasn\u0027t nearly as rare as I\u0027d hoped it to be. In fact its quite\ncommon - and given it walks the whole machine, thats very bad.\n\nThe new approach is simple (why didn\u0027t I think of it before?), we set the\naggregate shares to the full task group weight, and each larger sched domain\nthat encounters an aggregate shares larger than the weight, clips it (it\nalready re-distributes anyway).\n\nThis nicely converges to the desired global picture where the sum of all\nshares equals the task group weight.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "03970f065d9b4b156d0e879f82989440f7045396", "tree": "7b81e32b6730a682fa945136a71ed7a6f59f6640", "parents": [ "472613b961affef0c73f1c797993678312e7c666" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Tue Apr 22 10:04:26 2008 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Thu Apr 24 14:40:29 2008 -0700" }, "message": "[PATCH] Build fix for CONFIG_NUMA\u003dy \u0026\u0026 CONFIG_SMP\u003dn\n\nRegression caused by 434d53b00d6bb7be0a1d3dcc0d0d5df6c042e164\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "472613b961affef0c73f1c797993678312e7c666", "tree": "b350341d0f10ee150020697b81ebc900332cbba0", "parents": [ "94bc891b00e40cbec375feb4568780af183fd7f4" ], "author": { "name": "Russ Anderson", "email": "rja@sgi.com", "time": "Thu Apr 24 13:16:59 2008 -0500" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Thu Apr 24 14:21:21 2008 -0700" }, "message": "[IA64] fix bootmem regression on Altix\n\nA recent change prevents SGI Altix from booting.\nThis patch fixes the problem.\n\nThe regresson was introduced in commit 434d53b00d6bb7be0a1d3dcc0d0d5df6c042e164\n\nSigned-off-by: Russ Anderson \u003crja@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "73486722b70a83bba17be722677519b0855abedf", "tree": "76da8b2e9028afb4fae26c48b07eb9e27a5810a4", "parents": [ "03b883840c630c7b571690577809754739ac773f" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Tue Apr 22 10:07:22 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 22 13:48:02 2008 -0700" }, "message": "kernel-doc: fix sched.c missing parameter\n\nAdd missing kernel-doc in kernel/sched.c:\n\nWarning(linux-2.6.25-git3//kernel/sched.c:7044): No description found for parameter \u0027span\u0027\n\nSigned-off-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c24b7c524421f9ea9d9ebab55f80cfb1f3fb77a3", "tree": "52a61a1091c8d119b5a2ae97661651d17c2e9d5e", "parents": [ "f00b45c145981b43c7e7f66315ac77534c938cbf" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Apr 18 10:55:34 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: features fix\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f00b45c145981b43c7e7f66315ac77534c938cbf", "tree": "56604b844199ded71c518706f3c3a11275432266", "parents": [ "06379aba522ebdabca37446ea988a23c43c03c67" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: /debug/sched_features\n\nprovide a text based interface to the scheduler features; this saves the\n\u0027user\u0027 from setting bits using decimal arithmetic.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "06379aba522ebdabca37446ea988a23c43c03c67", "tree": "7a6b522e6f8390935e5369b96d1a080a0ee579d7", "parents": [ "7ba2e74ab5a0518bc953042952dd165724bc70c9" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 09:25:58 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: add SCHED_FEAT_DEADLINE\n\nunused at the moment.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "8f1bc385cfbab474db6c27b5af1e439614f3025c", "tree": "e4cfb8255d62621d17bc08ee5f94f42a0cc22677", "parents": [ "4a55bd5e97b1775913f88f11108a4f144f590e89" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: fair: weight calculations\n\nIn order to level the hierarchy, we need to calculate load based on the\nroot view. That is, each task\u0027s load is in the same unit.\n\n A\n / \\\n B 1\n / \\\n 2 3\n\nTo compute 1\u0027s load we do:\n\n\t weight(1)\n\t--------------\n\t rq_weight(A)\n\nTo compute 2\u0027s load we do:\n\n\t weight(2) weight(B)\n\t------------ * -----------\n\trq_weight(B) rw_weight(A)\n\nThis yields load fractions in comparable units.\n\nThe consequence is that it changes virtual time. We used to have:\n\n time_{i}\n vtime_{i} \u003d ------------\n weight_{i}\n\n vtime \u003d \\Sum vtime_{i} \u003d time / rq_weight.\n\nBut with the new way of load calculation we get that vtime equals time.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4a55bd5e97b1775913f88f11108a4f144f590e89", "tree": "4514f2370d898b93086779c821023319fe4c8b9d", "parents": [ "ac884dec6d4a7df252150af875cffddf8f1d9c15" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: fair-group: de-couple load-balancing from the rb-trees\n\nDe-couple load-balancing from the rb-trees, so that I can change their\norganization.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "18d95a2832c1392a2d63227a7a6d433cb9f2037e", "tree": "fa85b700aa3caac5b1309edd8e31d9b957957a83", "parents": [ "1d3504fcf5606579d60b649d19f44b3871c1ddae" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: fair-group: SMP-nice for group scheduling\n\nImplement SMP nice support for the full group hierarchy.\n\nOn each load-balance action, compile a sched_domain wide view of the full\ntask_group tree. We compute the domain wide view when walking down the\nhierarchy, and readjust the weights when walking back up.\n\nAfter collecting and readjusting the domain wide view, we try to balance the\ntasks within the task_groups. The current approach is a naively balance each\ntask group until we\u0027ve moved the targeted amount of load.\n\nInspired by Srivatsa Vaddsgiri\u0027s previous code and Abhishek Chandra\u0027s H-SMP\npaper.\n\nXXX: there will be some numerical issues due to the limited nature of\n SCHED_LOAD_SCALE wrt to representing a task_groups influence on the\n total weight. When the tree is deep enough, or the task weight small\n enough, we\u0027ll run out of bits.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCC: Abhishek Chandra \u003cchandra@cs.umn.edu\u003e\nCC: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "1d3504fcf5606579d60b649d19f44b3871c1ddae", "tree": "001a1b57bd7f123fd51361d78d9277b2dcac1bf3", "parents": [ "4d5f35533fb9b2cd553cec6611195bcbfb7ffd84" ], "author": { "name": "Hidetoshi Seto", "email": "seto.hidetoshi@jp.fujitsu.com", "time": "Tue Apr 15 14:04:23 2008 +0900" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched, cpuset: customize sched domains, core\n\n[rebased for sched-devel/latest]\n\n - Add a new cpuset file, having levels:\n sched_relax_domain_level\n\n - Modify partition_sched_domains() and build_sched_domains()\n to take attributes parameter passed from cpuset.\n\n - Fill newidle_idx for node domains which currently unused but\n might be required if sched_relax_domain_level become higher.\n\n - We can change the default level by boot option \u0027relax_domain_level\u003d\u0027.\n\nSigned-off-by: Hidetoshi Seto \u003cseto.hidetoshi@jp.fujitsu.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "b40b2e8eb52192a8a22d707ed37925792b7bdfd1", "tree": "88a20a3fc5a3d80ad7f5db4cd26255e14fd82642", "parents": [ "f473aa5e025bc8e0c5fe9352f65178a54adadec2" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: rt: multi level group constraints\n\nmulti level rt constraints\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f473aa5e025bc8e0c5fe9352f65178a54adadec2", "tree": "4b5f2de42e619c4a6fa62c6bc20474d5d8c9b258", "parents": [ "eff766a65c60237bfa865160c3129de31fab591b" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: task_group hierarchy\n\nAdd the full parent\u003c-\u003echild relation thing into task_groups as well.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "eff766a65c60237bfa865160c3129de31fab591b", "tree": "c003be272be4bd614485606c77893295f3b175a2", "parents": [ "ec7dc8ac73e4a56ed03b673f026f08c0d547f597" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:45:00 2008 +0200" }, "message": "sched: fix the task_group hierarchy for UID grouping\n\nUID grouping doesn\u0027t actually have a task_group representing the root of\nthe task_group tree. Add one.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "ec7dc8ac73e4a56ed03b673f026f08c0d547f597", "tree": "851dd612f2c80e4643d10c18bb33ab5b2df9aa70", "parents": [ "354d60c2ff72d86627dfe2089d186824abf4bb8e" ], "author": { "name": "Dhaval Giani", "email": "dhaval@linux.vnet.ibm.com", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: allow the group scheduler to have multiple levels\n\nThis patch makes the group scheduler multi hierarchy aware.\n\n[a.p.zijlstra@chello.nl: rt-parts and assorted fixes]\nSigned-off-by: Dhaval Giani \u003cdhaval@linux.vnet.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "354d60c2ff72d86627dfe2089d186824abf4bb8e", "tree": "10cea61ce7036448ed7246820c5575df2a61bb3b", "parents": [ "ea736ed5d353d7a3aa1cf8ce4cf8d947bc353fb2" ], "author": { "name": "Dhaval Giani", "email": "dhaval@linux.vnet.ibm.com", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: mix tasks and groups\n\nThis patch allows tasks and groups to exist in the same cfs_rq. With this\nchange the CFS group scheduling follows a 1/(M+N) model from a 1/(1+N)\nfairness model where M tasks and N groups exist at the cfs_rq level.\n\n[a.p.zijlstra@chello.nl: rt bits and assorted fixes]\nSigned-off-by: Dhaval Giani \u003cdhaval@linux.vnet.ibm.com\u003e\nSigned-off-by: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "ea736ed5d353d7a3aa1cf8ce4cf8d947bc353fb2", "tree": "1675a12c641607c95c67969745dbfa12e7815cb8", "parents": [ "112f53f5d700589de741dca67c77439e96ea94a7" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Mar 25 13:51:45 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: fix checks\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "112f53f5d700589de741dca67c77439e96ea94a7", "tree": "1faeaa6a25717b80c8ce0a51a9bf7f5f32d5ac08", "parents": [ "cd8ba7cd9be0192348c2836cb6645d9b2cd2bfd2" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Mar 19 11:43:36 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: old sleeper bonus\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "cd8ba7cd9be0192348c2836cb6645d9b2cd2bfd2", "tree": "3b4138c7b683c2168ac13be41aab74b49a6bcf1c", "parents": [ "e0982e90cd1ecf59818b137386b7f63debded9cc" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Wed Mar 26 14:23:49 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: add new set_cpus_allowed_ptr function\n\nAdd a new function that accepts a pointer to the \"newly allowed cpus\"\ncpumask argument.\n\nint set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)\n\nThe current set_cpus_allowed() function is modified to use the above\nbut this does not result in an ABI change. And with some compiler\noptimization help, it may not introduce any additional overhead.\n\nAdditionally, to enforce the read only nature of the new_mask arg, the\n\"const\" property is migrated to sub-functions called by set_cpus_allowed.\nThis silences compiler warnings.\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "e0982e90cd1ecf59818b137386b7f63debded9cc", "tree": "3cdbfa8a69dca4a9c9596d61bffa32f1b676f09d", "parents": [ "4bdbaad33d0f4d0e9818a38a825f5b75c0296a28" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Wed Mar 26 14:23:48 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "init: move setup of nr_cpu_ids to as early as possible\n\nMove the setting of nr_cpu_ids from sched_init() to start_kernel()\nso that it\u0027s available as early as possible.\n\nNote that an arch has the option of setting it even earlier if need be,\nbut it should not result in a different value than the setup_nr_cpu_ids()\nfunction.\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4bdbaad33d0f4d0e9818a38a825f5b75c0296a28", "tree": "36d8ac15fb3d2d4ccb939327a786f8327a403b98", "parents": [ "9d1fe3236a1d64ab687e16b4cbbaa1383352a2c1" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Tue Apr 15 16:35:52 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "sched: remove another cpumask_t variable from stack\n\n * Remove another cpumask_t variable from stack that was missed in the\n last kernel_sched_c updates.\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "7c16ec585c558960a508ccf9a08fcb9ed49b3754", "tree": "cca2b12203a10944d7095a07df7292421f578dc9", "parents": [ "c5f59f0833df945eef7ff35f3dc6ba61c5f293dd" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Fri Apr 04 18:11:11 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "cpumask: reduce stack usage in SD_x_INIT initializers\n\n * Remove empty cpumask_t (and all non-zero/non-null) variables\n in SD_*_INIT macros. Use memset(0) to clear. Also, don\u0027t\n inline the initializer functions to save on stack space in\n build_sched_domains().\n\n * Merge change to include/linux/topology.h that uses the new\n node_to_cpumask_ptr function in the nr_cpus_node macro into\n this patch.\n\nDepends on:\n\t[mm-patch]: asm-generic-add-node_to_cpumask_ptr-macro.patch\n\t[sched-devel]: sched: add new set_cpus_allowed_ptr function\n\nCc: H. Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "c5f59f0833df945eef7ff35f3dc6ba61c5f293dd", "tree": "32c1a94847d0154051c79011212d401462723d55", "parents": [ "b53e921ba1cff8453dc9a87a84052fa12d5b30bd" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Fri Apr 04 18:11:10 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "nodemask: use new node_to_cpumask_ptr function\n\n * Use new node_to_cpumask_ptr. This creates a pointer to the\n cpumask for a given node. This definition is in mm patch:\n\n\tasm-generic-add-node_to_cpumask_ptr-macro.patch\n\n * Use new set_cpus_allowed_ptr function.\n\nDepends on:\n\t[mm-patch]: asm-generic-add-node_to_cpumask_ptr-macro.patch\n\t[sched-devel]: sched: add new set_cpus_allowed_ptr function\n\t[x86/latest]: x86: add cpus_scnprintf function\n\nCc: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\nCc: Greg Banks \u003cgnb@melbourne.sgi.com\u003e\nCc: H. Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "b53e921ba1cff8453dc9a87a84052fa12d5b30bd", "tree": "021cadb6c58543ecccd95b55fc319f249ebf176e", "parents": [ "f9a86fcbbb1e5542eabf45c9144ac4b6330861a4" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Fri Apr 04 18:11:08 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:59 2008 +0200" }, "message": "generic: reduce stack pressure in sched_affinity\n\n * Modify sched_affinity functions to pass cpumask_t variables by reference\n instead of by value.\n\n * Use new set_cpus_allowed_ptr function.\n\nDepends on:\n\t[sched-devel]: sched: add new set_cpus_allowed_ptr function\n\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Cliff Wickman \u003ccpw@sgi.com\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f9a86fcbbb1e5542eabf45c9144ac4b6330861a4", "tree": "0a3f8d57969b2dc8d2663e05d6ee36f9b50ba26a", "parents": [ "f70316dace2bb99730800d47044acb818c6735f6" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Fri Apr 04 18:11:07 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "message": "cpuset: modify cpuset_set_cpus_allowed to use cpumask pointer\n\n * Modify cpuset_cpus_allowed to return the currently allowed cpuset\n via a pointer argument instead of as the function return value.\n\n * Use new set_cpus_allowed_ptr function.\n\n * Cleanup CPU_MASK_ALL and NODE_MASK_ALL uses.\n\nDepends on:\n\t[sched-devel]: sched: add new set_cpus_allowed_ptr function\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "434d53b00d6bb7be0a1d3dcc0d0d5df6c042e164", "tree": "9bf697c0891127157e75fea409d8d8167d210fd4", "parents": [ "d366f8cbc16882e93538d9a52423c2f50dad7c06" ], "author": { "name": "Mike Travis", "email": "travis@sgi.com", "time": "Fri Apr 04 18:11:04 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "message": "sched: remove fixed NR_CPUS sized arrays in kernel_sched_c\n\n * Change fixed size arrays to per_cpu variables or dynamically allocated\n arrays in sched_init() and sched_init_smp().\n\n (1) static struct sched_entity *init_sched_entity_p[NR_CPUS];\n (1) static struct cfs_rq *init_cfs_rq_p[NR_CPUS];\n (1) static struct sched_rt_entity *init_sched_rt_entity_p[NR_CPUS];\n (1) static struct rt_rq *init_rt_rq_p[NR_CPUS];\n\t static struct sched_group **sched_group_nodes_bycpu[NR_CPUS];\n\n (1) - these arrays are allocated via alloc_bootmem_low()\n\n * Change sched_domain_debug_one() to use cpulist_scnprintf instead of\n cpumask_scnprintf. This reduces the output buffer required and improves\n readability when large NR_CPU count machines arrive.\n\n * In sched_create_group() we allocate new arrays based on nr_cpu_ids.\n\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "0297b80339d545045490716fa8591b215fdd9458", "tree": "3a00c5bd83fc3f0493a2a26b95db0c4a7e4bb691", "parents": [ "32cd756a80aaef657ac09c76e6eff3ba65567790" ], "author": { "name": "Dhaval Giani", "email": "dhaval@linux.vnet.ibm.com", "time": "Fri Feb 29 10:02:44 2008 +0530" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "message": "sched: allow cpuacct stats to be reset\n\nCurrently the schedstats implementation does not allow the statistics\nto be reset. This patch aims to allow that.\n\n echo 0 \u003e cpuacct.usage\n\nresets the usage. Any other value is not allowed and returns -EINVAL.\n\nSigned-off-by: Dhaval Giani \u003cdhaval@linux.vnet.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "32cd756a80aaef657ac09c76e6eff3ba65567790", "tree": "96945c0d1ffb03694757d174ba5969bead89aa31", "parents": [ "48f20a9a9488c432fc86df1ff4b7f4fa895d1183" ], "author": { "name": "Dhaval Giani", "email": "dhaval@linux.vnet.ibm.com", "time": "Fri Feb 29 10:02:43 2008 +0530" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "message": "sched: cleanup cpuacct variable names\n\nChange the variable names to the common convention for the cpuacct\nsubsystem.\n\nSigned-off-by: Dhaval Giani \u003cdhaval@linux.vnet.ibm.com\u003e\nAcked-by: Balbir Singh \u003cbalbir@linux.vnet.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "ac086bc22997a2be24fc40fc8d46522fe7e03d11", "tree": "7a484ba13acbdf0fa98c896ce58e807b4b5b1af9", "parents": [ "d0b27fa77854b149ad4af08b0fe47fe712a47ade" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:58 2008 +0200" }, "message": "sched: rt-group: smp balancing\n\nCurrently the rt group scheduling does a per cpu runtime limit, however\nthe rt load balancer makes no guarantees about an equal spread of real-\ntime tasks, just that at any one time, the highest priority tasks run.\n\nSolve this by making the runtime limit a global property by borrowing\nexcessive runtime from the other cpus once the local limit runs out.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d0b27fa77854b149ad4af08b0fe47fe712a47ade", "tree": "2f4487c108a5132e8d52456567b4a67e78fbb4a6", "parents": [ "57d3da2911787a101a384532f4519f9640bae883" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: rt-group: synchonised bandwidth period\n\nVarious SMP balancing algorithms require that the bandwidth period\nrun in sync.\n\nPossible improvements are moving the rt_bandwidth thing into root_domain\nand keeping a span per rt_bandwidth which marks throttled cpus.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "50df5d6aea6694ca481b8005900401e8c95c2603", "tree": "3b2e2955aceed795f5d54220ee37a859e83c12c6", "parents": [ "02e2b83bd25bb05ac2e69cb31458b7d1b3c70707" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 14 16:09:59 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: remove sysctl_sched_batch_wakeup_granularity\n\nit\u0027s unused.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "02e2b83bd25bb05ac2e69cb31458b7d1b3c70707", "tree": "f902ed65b94929aa78509150f52a26661974fac6", "parents": [ "d25ce4cd499a21aab89ff8755f8c4a2800eae25f" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 01:37:10 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: reenable sync wakeups\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d25ce4cd499a21aab89ff8755f8c4a2800eae25f", "tree": "a70f58303a4b1fabfffbf097188a6c67a888ee0e", "parents": [ "1fc8afa4c820fcde3658238eab5c010476ede521" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Mar 17 09:36:53 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: cache hot buddy\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "1fc8afa4c820fcde3658238eab5c010476ede521", "tree": "6a434a569bb8158e89d071c6d87a42f0fde3f697", "parents": [ "b85d0667268320072ccdeb07c27c25b300ab3724" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 01:39:19 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: feat affine wakeups\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "b85d0667268320072ccdeb07c27c25b300ab3724", "tree": "34bc3926e6524d82d450df5531744fff6b4df560", "parents": [ "0bbd3336eee1e712a290e0dfd1a64cbbdd63a508" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Mar 16 20:03:22 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: introduce SCHED_FEAT_SYNC_WAKEUPS, turn it off\n\nturn off sync wakeups by default. They are not needed anymore - the\nbuddy logic should be smart enough to keep the system from\noverscheduling.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "15934a37324f32e0fda633dc7984a671ea81cd75", "tree": "1f65ac7d910e76b65d0cfcd1c72e156b0a8bd273", "parents": [ "30914a58af9d21c5f1831adabb5d7a800a378675" ], "author": { "name": "Guillaume Chazarain", "email": "guichaz@yahoo.fr", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: fix rq-\u003eclock overflows detection with CONFIG_NO_HZ\n\nWhen using CONFIG_NO_HZ, rq-\u003etick_timestamp is not updated every TICK_NSEC.\nWe check that the number of skipped ticks matches the clock jump seen in\n__update_rq_clock().\n\nSigned-off-by: Guillaume Chazarain \u003cguichaz@yahoo.fr\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "30914a58af9d21c5f1831adabb5d7a800a378675", "tree": "8e63c0974c94a7e7106160274995f89717991722", "parents": [ "27ec4407790d075c325e1f4da0a19c56953cce23" ], "author": { "name": "Reynes Philippe", "email": "tremyfr@yahoo.fr", "time": "Mon Mar 17 16:19:05 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: sched.c needs tick.h\n\nkernel/sched.c:506: erreur: implicit declaration of function tick_get_tick_sched\nkernel/sched.c:506: erreur: invalid type argument of -\u003e\nkernel/sched.c:506: erreur: NOHZ_MODE_INACTIVE undeclared (first use in this function)\nkernel/sched.c:506: erreur: (Each undeclared identifier is reported only once\nkernel/sched.c:506: erreur: for each function it appears in.)\n\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "27ec4407790d075c325e1f4da0a19c56953cce23", "tree": "d8c068c4743394eab173995bf085a99dfec257c7", "parents": [ "018d6db4cb5bbdcd65424a16f2dcca692ed32ae4" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Feb 28 21:00:21 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Apr 19 19:44:57 2008 +0200" }, "message": "sched: make cpu_clock() globally synchronous\n\nAlexey Zaytsev reported (and bisected) that the introduction of\ncpu_clock() in printk made the timestamps jump back and forth.\n\nMake cpu_clock() more reliable while still keeping it fast when it\u0027s\ncalled frequently.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "06d8308c61e54346585b2691c13ee3f90cb6fb2f", "tree": "d522e285525964706e7ec112bca23d407a540a8e", "parents": [ "898a19de1502649877091b398229026b4142c0e2" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sat Mar 22 09:20:24 2008 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Wed Mar 26 08:28:55 2008 +0100" }, "message": "NOHZ: reevaluate idle sleep length after add_timer_on()\n\nadd_timer_on() can add a timer on a CPU which is currently in a long\nidle sleep, but the timer wheel is not reevaluated by the nohz code on\nthat CPU. So a timer can be delayed for quite a long time. This\ntriggered a false positive in the clocksource watchdog code.\n\nTo avoid this we need to wake up the idle CPU and enforce the\nreevaluation of the timer wheel for the next timer event.\n\nAdd a function, which checks a given CPU for idle state, marks the\nidle task with NEED_RESCHED and sends a reschedule IPI to notify the\nother CPU of the change in the timer wheel.\n\nCall this function from add_timer_on().\n\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: stable@kernel.org\n\n--\n include/linux/sched.h | 6 ++++++\n kernel/sched.c | 43 +++++++++++++++++++++++++++++++++++++++++++\n kernel/timer.c | 10 +++++++++-\n 3 files changed, 58 insertions(+), 1 deletion(-)\n\n" }, { "commit": "22e52b072dd87faa9b2559fe89d4e8f2370f81ca", "tree": "79e71d0aad3c9ea9f326977c3a3ff1dceaec81b5", "parents": [ "9aefd0abd8610e8f3bb097debf3afb73f8b7b210" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Mar 12 18:31:59 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 21 16:43:48 2008 +0100" }, "message": "sched: add arch_update_cpu_topology hook.\n\nWill be called each time the scheduling domains are rebuild.\nNeeded for architectures that don\u0027t have a static cpu topology.\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "9aefd0abd8610e8f3bb097debf3afb73f8b7b210", "tree": "3456c6949cd8b9c926aa9beef7c9bf99d0a84caa", "parents": [ "23e3c3cd2e39a3c9d07ee07d882c8cf6ddd61c86" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Mar 12 18:31:58 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 21 16:43:47 2008 +0100" }, "message": "sched: add exported arch_reinit_sched_domains() to header file.\n\nNeeded so it can be called from outside of sched.c.\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "23e3c3cd2e39a3c9d07ee07d882c8cf6ddd61c86", "tree": "a728df75b68d5648ea6d6627125c48f6d56fe1a8", "parents": [ "2070ee01d314ecec8a570c07647ccf4ced6340bb" ], "author": { "name": "Roel Kluin", "email": "12o3l@tiscali.nl", "time": "Thu Mar 13 17:41:59 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 21 16:43:47 2008 +0100" }, "message": "sched: remove double unlikely from schedule()\n\nCombine two unlikely\u0027s\n\nSigned-off-by: Roel Kluin \u003c12o3l@tiscali.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "2070ee01d314ecec8a570c07647ccf4ced6340bb", "tree": "e7a3c95e9c8ae06917e00a40fcb6cae1c4a9c924", "parents": [ "ae51801ba5ca27c2c571eb508daa99b392e79bd4" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Mar 21 16:43:47 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 21 16:43:47 2008 +0100" }, "message": "sched: cleanup old and rarely used \u0027debug\u0027 features.\n\nTREE_AVG and APPROX_AVG are initial task placement policies that have been\ndisabled for a long while.. time to remove them.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCC: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f540a6080a092e2ab69fd146c308022db7347b0a", "tree": "12dda1ecd63adb17cf06db61712a113cf6990037", "parents": [ "4ae7d5cefd4aa3560e359a3b0f03e12adc8b5c86" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Mar 15 17:10:34 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 04:27:53 2008 +0100" }, "message": "sched: wakeup-buddy tasks are cache-hot\n\nWakeup-buddy tasks are cache-hot - this makes it a bit harder\nfor the load-balancer to tear them apart. (but it\u0027s still possible,\nif the load is sufficiently assymetric)\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4ae7d5cefd4aa3560e359a3b0f03e12adc8b5c86", "tree": "c7517529ec199d3d9936b4049880a820534d2703", "parents": [ "6f3d09291b4982991680b61763b2541e53e2a95f" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 01:42:00 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 04:27:53 2008 +0100" }, "message": "sched: improve affine wakeups\n\nimprove affine wakeups. Maintain the \u0027overlap\u0027 metric based on CFS\u0027s\nsum_exec_runtime - which means the amount of time a task executes\nafter it wakes up some other task.\n\nUse the \u0027overlap\u0027 for the wakeup decisions: if the \u0027overlap\u0027 is short,\nit means there\u0027s strong workload coupling between this task and the\nwoken up task. If the \u0027overlap\u0027 is large then the workload is decoupled\nand the scheduler will move them to separate CPUs more easily.\n\n( Also slightly move the preempt_check within try_to_wake_up() - this has\n no effect on functionality but allows \u0027early wakeups\u0027 (for still-on-rq\n tasks) to be correctly accounted as well.)\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "aa2ac25229cd4d0280f6174c42712744ad61b140", "tree": "e4450de1bb2cd4cd56d6abf64feb862c1d542653", "parents": [ "27d117266097101dcf79c4576903cdcdd0eabffc" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Mar 14 21:12:12 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Mar 15 03:02:50 2008 +0100" }, "message": "sched: fix overload performance: buddy wakeups\n\nCurrently we schedule to the leftmost task in the runqueue. When the\nruntimes are very short because of some server/client ping-pong,\nespecially in over-saturated workloads, this will cycle through all\ntasks trashing the cache.\n\nReduce cache trashing by keeping dependent tasks together by running\nnewly woken tasks first. However, by not running the leftmost task first\nwe could starve tasks because the wakee can gain unlimited runtime.\n\nTherefore we only run the wakee if its within a small\n(wakeup_granularity) window of the leftmost task. This preserves\nfairness, but does alternate server/client task groups.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "27d117266097101dcf79c4576903cdcdd0eabffc", "tree": "854e6e824b03385d71d8d7f2feb9311572bb86f2", "parents": [ "e89996ae3f9e88d4fd75751a15c10b19d197e702" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 14 22:20:01 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Mar 15 03:02:50 2008 +0100" }, "message": "sched: fix calc_delta_mine()\n\nlw-\u003eweight can be 0 for a short time during bootup.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n" }, { "commit": "e89996ae3f9e88d4fd75751a15c10b19d197e702", "tree": "b4b754697995e3ebff4e987b46167263e86ff0d8", "parents": [ "3fe69747dab906cd6a8523230276a9820d6a514f" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 14 23:48:28 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Mar 15 03:02:49 2008 +0100" }, "message": "sched: fix update_load_add()/sub()\n\nClear the cached inverse value when updating load. This is needed for\ncalc_delta_mine() to work correctly when using the rq load.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n" }, { "commit": "0e1f34833bd9170ccc93ab759e48e695917fa48f", "tree": "2ca4b89dd755f2d82dbd34466a8cd6d8b5b4a0af", "parents": [ "4faa8496650f9417189dacce8f933e8ec61dc032" ], "author": { "name": "Hiroshi Shimamoto", "email": "h-shimamoto@ct.jp.nec.com", "time": "Mon Mar 10 11:01:20 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sat Mar 15 03:02:49 2008 +0100" }, "message": "sched: fix race in schedule()\n\nFix a hard to trigger crash seen in the -rt kernel that also affects\nthe vanilla scheduler.\n\nThere is a race condition between schedule() and some dequeue/enqueue\nfunctions; rt_mutex_setprio(), __setscheduler() and sched_move_task().\n\nWhen scheduling to idle, idle_balance() is called to pull tasks from\nother busy processor. It might drop the rq lock. It means that those 3\nfunctions encounter on_rq\u003d0 and running\u003d1. The current task should be\nput when running.\n\nHere is a possible scenario:\n\n CPU0 CPU1\n | schedule()\n | -\u003edeactivate_task()\n | -\u003eidle_balance()\n | --\u003eload_balance_newidle()\nrt_mutex_setprio() |\n | ---\u003edouble_lock_balance()\n *get lock *rel lock\n * on_rq\u003d0, ruuning\u003d1 |\n * sched_class is changed |\n *rel lock *get lock\n : |\n :\n -\u003eput_prev_task_rt()\n -\u003epick_next_task_fair()\n \u003d\u003e panic\n\nThe current process of CPU1(P1) is scheduling. Deactivated P1, and the\nscheduler looks for another process on other CPU\u0027s runqueue because CPU1\nwill be idle. idle_balance(), load_balance_newidle() and\ndouble_lock_balance() are called and double_lock_balance() could drop\nthe rq lock. On the other hand, CPU0 is trying to boost the priority of\nP1. The result of boosting only P1\u0027s prio and sched_class are changed to\nRT. The sched entities of P1 and P1\u0027s group are never put. It makes\ncfs_rq invalid, because the cfs_rq has curr and no leaf, but\npick_next_task_fair() is called, then the kernel panics.\n\nSigned-off-by: Hiroshi Shimamoto \u003ch-shimamoto@ct.jp.nec.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "08f503b0c089968b2542659a89dfd50c5c59bb0b", "tree": "0a88cc425c2c213a2e6384f4df0739e2e8485e8f", "parents": [ "1f94ef598e8d29b92b9fc85d43c832e03721d3cb" ], "author": { "name": "Gregory Haskins", "email": "ghaskins@novell.com", "time": "Mon Mar 10 17:59:11 2008 -0400" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Mar 11 14:02:58 2008 +0100" }, "message": "keep rd-\u003eonline and cpu_online_map in sync\n\nIt is possible to allow the root-domain cache of online cpus to\nbecome out of sync with the global cpu_online_map. This is because we\ncurrently trigger removal of cpus too early in the notifier chain.\nOther DOWN_PREPARE handlers may in fact run and reconfigure the\nroot-domain topology, thereby stomping on our own offline handling.\n\nThe end result is that rd-\u003eonline may become out of sync with\ncpu_online_map, which results in potential task misrouting.\n\nSo change the offline handling to be more tightly coupled with the\nglobal offline process by triggering on CPU_DYING intead of\nCPU_DOWN_PREPARE.\n\nSigned-off-by: Gregory Haskins \u003cghaskins@novell.com\u003e\nCc: Gautham R Shenoy \u003cego@in.ibm.com\u003e\nCc: \"Siddha, Suresh B\" \u003csuresh.b.siddha@intel.com\u003e\nCc: \"Rafael J. Wysocki\" \u003crjw@sisk.pl\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "1f94ef598e8d29b92b9fc85d43c832e03721d3cb", "tree": "f9573663c4bb83a9af4d17b90c174869b50ba2b1", "parents": [ "2f44bbb495dd3e6d0209eff2257438ab9c570e5b" ], "author": { "name": "Gregory Haskins", "email": "ghaskins@novell.com", "time": "Mon Mar 10 16:52:41 2008 -0400" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Mar 11 14:02:58 2008 +0100" }, "message": "Revert \"cpu hotplug: adjust root-domain-\u003eonline span in response to hotplug event\"\n\nThis reverts commit 393d94d98b19089ec172566e23557997931b137e.\n\nLets fix this right.\n\nSigned-off-by: Gregory Haskins \u003cghaskins@novell.com\u003e\nCc: Gautham R Shenoy \u003cego@in.ibm.com\u003e\nCc: \"Siddha, Suresh B\" \u003csuresh.b.siddha@intel.com\u003e\nCc: \"Rafael J. Wysocki\" \u003crjw@sisk.pl\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "393d94d98b19089ec172566e23557997931b137e", "tree": "62154269fd93690ffc07ad885cf371542b04ecc5", "parents": [ "6efcae460186c0c1c94afff58a92784e1fc0d10b" ], "author": { "name": "Gregory Haskins", "email": "ghaskins@novell.com", "time": "Sat Mar 08 00:10:15 2008 -0500" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Mar 09 10:05:14 2008 -0700" }, "message": "cpu hotplug: adjust root-domain-\u003eonline span in response to hotplug event\n\nWe currently set the root-domain online span automatically when the\ndomain is added to the cpu if the cpu is already a member of\ncpu_online_map.\n\nThis was done as a hack/bug-fix for s2ram, but it also causes a problem\nwith hotplug CPU_DOWN transitioning. The right way to fix the original\nproblem is to actually respond to CPU_UP events, instead of CPU_ONLINE,\nwhich is already too late.\n\nThis solves the hung reboot regression reported by Andrew Morton and\nothers.\n\nSigned-off-by: Gregory Haskins \u003cghaskins@novell.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "521f1a2489c41f8b1181b0a8eb52e1c34284d50b", "tree": "cfe339df98dcb65aaae6c45021ac92cdf60c6540", "parents": [ "2692a2406b9262bbb101708815be99ec2988e48b" ], "author": { "name": "Dhaval Giani", "email": "dhaval@linux.vnet.ibm.com", "time": "Thu Feb 28 15:21:56 2008 +0530" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 07 16:43:00 2008 +0100" }, "message": "sched: don\u0027t allow rt_runtime_us to be zero for groups having rt tasks\n\nThis patch checks if we can set the rt_runtime_us to 0. If there is a\nrealtime task in the group, we don\u0027t want to set the rt_runtime_us as 0\nor bad things will happen. (that task wont get any CPU time despite\nbeing TASK_RUNNNG)\n\nSigned-off-by: Dhaval Giani \u003cdhaval@linux.vnet.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "2692a2406b9262bbb101708815be99ec2988e48b", "tree": "5f956c6f28e7e83f5a1c57e2724b1739da25da30", "parents": [ "1868f958eb56fc41c5985c8732e564a400c5fdf5" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 27 12:00:46 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 07 16:43:00 2008 +0100" }, "message": "sched: rt-group: fixup schedulability constraints calculation\n\nit was only possible to configure the rt-group scheduling parameters\nbeyond the default value in a very small range.\n\nthat\u0027s because div64_64() has a different calling convention than\ndo_div() :/\n\nfix a few untidies while we are here; sysctl_sched_rt_period may overflow\ndue to that multiplication, so cast to u64 first. Also that RUNTIME_INF\njuggling makes little sense although its an effective NOP.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "1868f958eb56fc41c5985c8732e564a400c5fdf5", "tree": "4f8d83e487e5fa93e3d773591a26eda02d1b387d", "parents": [ "150d8bede7f85eb00d8f4d628e6b0bae68739e3b" ], "author": { "name": "Miao Xie", "email": "miaox@cn.fujitsu.com", "time": "Fri Mar 07 09:35:06 2008 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 07 16:43:00 2008 +0100" }, "message": "sched: fix the wrong time slice value for SCHED_FIFO tasks\n\nFunction sys_sched_rr_get_interval returns wrong time slice value for\nSCHED_FIFO tasks. The time slice for SCHED_FIFO tasks should be 0.\n\nSigned-off-by: Miao Xie \u003cmiaox@cn.fujitsu.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "150d8bede7f85eb00d8f4d628e6b0bae68739e3b", "tree": "aa740c10b82ca59bd983713ef19569d9538b0373", "parents": [ "6fa46fa526f2cab9ce21fa5e39501553a40d196d" ], "author": { "name": "Pavel Roskin", "email": "proski@gnu.org", "time": "Wed Mar 05 16:56:37 2008 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 07 16:43:00 2008 +0100" }, "message": "sched: export task_nice\n\nThe API is trivial, and so is the implementation.\n\nSigned-off-by: Pavel Roskin \u003cproski@gnu.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "810b38179e9e4d4f57b4b733767bb08f8291a965", "tree": "91692de4fbd46879b29d803839b5bf7e25f30cd8", "parents": [ "ce4796d1e16cf3761dc2a02b8d588667d05b3078" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Feb 29 15:21:01 2008 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Mar 07 16:42:59 2008 +0100" }, "message": "sched: retain vruntime\n\nKei Tokunaga reported an interactivity problem when moving tasks\nbetween control groups.\n\nTasks would retain their old vruntime when moved between groups, this\ncan cause funny lags. Re-set the vruntime on group move to fit within\nthe new tree.\n\nReported-by: Kei Tokunaga \u003ctokunaga.keiich@jp.fujitsu.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "62fb185130e4d420f71a30ff59d8b16b74ef5d2b", "tree": "474c0824a5bf90950b0a430a11a52b358c9e1f31", "parents": [ "976dde010e513a9c7c3117a32b7b015f84b37430" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Mon Feb 25 17:34:02 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Mar 04 17:54:06 2008 +0100" }, "message": "sched: revert load_balance_monitor() changes\n\nThe following commits cause a number of regressions:\n\n commit 58e2d4ca581167c2a079f4ee02be2f0bc52e8729\n Author: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\n Date: Fri Jan 25 21:08:00 2008 +0100\n sched: group scheduling, change how cpu load is calculated\n\n commit 6b2d7700266b9402e12824e11e0099ae6a4a6a79\n Author: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\n Date: Fri Jan 25 21:08:00 2008 +0100\n sched: group scheduler, fix fairness of cpu bandwidth allocation for task groups\n\nNamely:\n - very frequent wakeups on SMP, reported by PowerTop users.\n - cacheline trashing on (large) SMP\n - some latencies larger than 500ms\n\nWhile there is a mergeable patch to fix the latter, the former issues\nare not fixable in a manner suitable for .25 (we\u0027re at -rc3 now).\n\nHence we revert them and try again in v2.6.26.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCC: Srivatsa Vaddagiri \u003cvatsa@linux.vnet.ibm.com\u003e\nTested-by: Alexey Zaytsev \u003calexey.zaytsev@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "67ca7bde2e9d3516b5ae0188330ad1059ac03f38", "tree": "770b1a1c7ed520d984b5de3c588c2e96368daafa", "parents": [ "7eee3e677d6e2e9007afcd7d79b0715525aa552e" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Fri Feb 15 09:56:36 2008 -0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Feb 25 16:34:17 2008 +0100" }, "message": "sched: fix signedness warnings in sched.c\n\nUnsigned long values are always assigned to switch_count,\nmake it unsigned long.\n\nkernel/sched.c:3897:15: warning: incorrect type in assignment (different signedness)\nkernel/sched.c:3897:15: expected long *switch_count\nkernel/sched.c:3897:15: got unsigned long *\u003cnoident\u003e\nkernel/sched.c:3921:16: warning: incorrect type in assignment (different signedness)\nkernel/sched.c:3921:16: expected long *switch_count\nkernel/sched.c:3921:16: got unsigned long *\u003cnoident\u003e\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "6892b75e60557a48c01d57ba320419a9e2ce9846", "tree": "2ba5c725d4723385e88b0a54e8bb4be5b9e38384", "parents": [ "bfa274e2436fc7ef72ef51c878083647f1cfd429" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 14:02:36 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Feb 25 16:34:16 2008 +0100" }, "message": "sched: make early bootup sched_clock() use safer\n\ndo not call sched_clock() too early. Not only might rq-\u003eidle\nnot be set up - but pure per-cpu data might not be accessible\neither.\n\nthis solves an ia64 early bootup hang with CONFIG_PRINTK_TIME\u003dy.\n\nTested-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nAcked-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nAcked-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "04e2f1741d235ba599037734878d72e57cb302b5", "tree": "5d4b05f9645370a40ce436aa6da18dc0b885d19c", "parents": [ "0a3716eb04ccfdbef6e872a343ba7ce309237e79" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sat Feb 23 18:05:03 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sat Feb 23 18:05:03 2008 -0800" }, "message": "Add memory barrier semantics to wake_up() \u0026 co\n\nOleg Nesterov and others have pointed out that on some architectures,\nthe traditional sequence of\n\n\tset_current_state(TASK_INTERRUPTIBLE);\n\tif (CONDITION)\n\t\treturn;\n\tschedule();\n\nis racy wrt another CPU doing\n\n\tCONDITION \u003d 1;\n\twake_up_process(p);\n\nbecause while set_current_state() has a memory barrier separating\nsetting of the TASK_INTERRUPTIBLE state from reading of the CONDITION\nvariable, there is no such memory barrier on the wakeup side.\n\nNow, wake_up_process() does actually take a spinlock before it reads and\nsets the task state on the waking side, and on x86 (and many other\narchitectures) that spinlock is in fact equivalent to a memory barrier,\nbut that is not generally guaranteed. The write that sets CONDITION\ncould move into the critical region protected by the runqueue spinlock.\n\nHowever, adding a smp_wmb() to before the spinlock should now order the\nwriting of CONDITION wrt the lock itself, which in turn is ordered wrt\nthe accesses within the spinlock (which includes the reading of the old\nstate).\n\nThis should thus close the race (which probably has never been seen in\npractice, but since smp_wmb() is a no-op on x86, it\u0027s not like this will\nmake anything worse either on the most common architecture where the\nspinlock already gave the required protection).\n\nAcked-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: Dmitry Adamushko \u003cdmitry.adamushko@gmail.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "43627582799db317e966ecb0002c2c3c9805ec0f", "tree": "fcfd50780b438d0ec04830faa857de9b86665e76", "parents": [ "0835ab53eab5bbeebe1c135e92fb0c4d483dde17" ], "author": { "name": "Srinivasa Ds", "email": "srinivasa@in.ibm.com", "time": "Sat Feb 23 15:24:04 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sat Feb 23 17:13:24 2008 -0800" }, "message": "kprobes: refuse kprobe insertion on add/sub_preempt_counter()\n\nKprobes makes use of preempt_disable(),preempt_enable_noresched() and these\nfunctions inturn call add/sub_preempt_count(). So we need to refuse user from\ninserting probe in to these functions.\n\nThis patch disallows user from probing add/sub_preempt_count().\n\nSigned-off-by: Srinivasa DS \u003csrinivasa@in.ibm.com\u003e\nAcked-by: Ananth N Mavinakayanahalli \u003cananth@in.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": "b68aa2300cabeb96801369a4bb37a4f19f59ed84", "tree": "f596af1a8d6da83982a1203b1021f4af7ed1f3f5", "parents": [ "bccbe08a60973c873e6af6fdb9ec11ffb1a6e4de" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "message": "sched: rt-group: refure unrunnable tasks\n\nRefuse to accept or create RT tasks in groups that can\u0027t run them.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "bccbe08a60973c873e6af6fdb9ec11ffb1a6e4de", "tree": "c0bc3eb67de4cb85f0de7d5b9c699ed5cc9386ff", "parents": [ "052f1dc7eb02300b05170ae341ccd03b76207778" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "message": "sched: rt-group: clean up the ifdeffery\n\nClean up some of the excessive ifdeffery introduces in the last patch.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "052f1dc7eb02300b05170ae341ccd03b76207778", "tree": "f58630b7876ae9e4308c0577e36aa13318b7bcfc", "parents": [ "9f0c1e560c43327b70998e6c702b2f01321130d9" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:40 2008 +0100" }, "message": "sched: rt-group: make rt groups scheduling configurable\n\nMake the rt group scheduler compile time configurable.\nKeep it experimental for now.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "9f0c1e560c43327b70998e6c702b2f01321130d9", "tree": "b2cc7ef5bb0dc9d7d2912de339bff3e0db3530c9", "parents": [ "23b0fdfc9299b137bd126e9dc22f62a59dae546d" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "message": "sched: rt-group: interface\n\nChange the rt_ratio interface to rt_runtime_us, to match rt_period_us.\nThis avoids picking a granularity for the ratio.\n\nExtend the /sys/kernel/uids/\u003cuid\u003e/ interface to allow setting\nthe group\u0027s rt_runtime.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "23b0fdfc9299b137bd126e9dc22f62a59dae546d", "tree": "22019172c555109b69a73da76561d99d7776c4f7", "parents": [ "4cf5d77a6eefaa7a464bc34e8cb767356f10fd74" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "message": "sched: rt-group: deal with PI\n\nSteven mentioned the fun case where a lock holding task will be throttled.\n\nSimple fix: allow groups that have boosted tasks to run anyway.\n\nIf a runnable task in a throttled group gets boosted the dequeue/enqueue\ndone by rt_mutex_setprio() is enough to unthrottle the group.\n\nThis is ofcourse not quite correct. Two possible ways forward are:\n - second prio array for boosted tasks\n - boost to a prio ceiling (this would also work for deadline scheduling)\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "4cf5d77a6eefaa7a464bc34e8cb767356f10fd74", "tree": "ad011b73207b6c8523189b873d8b1c3757d84e12", "parents": [ "8ed3699682be75fd68281239c202ad3830f9c72d" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "message": "sched: fix incorrect irq lock usage in normalize_rt_tasks()\n\nlockdep spotted this bogus irq locking. normalize_rt_tasks() can be called\nfrom hardirq context through sysrq-n\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "8ed3699682be75fd68281239c202ad3830f9c72d", "tree": "5af86fb155e475e906e14dc6aa5aa5a890cca7a2", "parents": [ "720a2592cf1b9af86f30c44e8d89348826c03372" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 13 15:45:39 2008 +0100" }, "message": "sched: fair-group: separate tg-\u003eshares from task_group_lock\n\nOn Mon, 2008-02-11 at 15:09 +0300, Denis V. Lunev wrote:\n\u003e BUG: sleeping function called from invalid context\n\u003e at /home/den/src/linux-netns26/kernel/mutex.c:209\n\u003e in_atomic():1, irqs_disabled():0\n\u003e no locks held by swapper/0.\n\u003e Pid: 0, comm: swapper Not tainted 2.6.24 #304\n\u003e\n\u003e Call Trace:\n\u003e \u003cIRQ\u003e [\u003cffffffff80252d1e\u003e] ? __debug_show_held_locks+0x15/0x27\n\u003e [\u003cffffffff8022c2a8\u003e] __might_sleep+0xc0/0xdf\n\u003e [\u003cffffffff8049f1df\u003e] mutex_lock_nested+0x28/0x2a9\n\u003e [\u003cffffffff80231294\u003e] sched_destroy_group+0x18/0xea\n\u003e [\u003cffffffff8023e835\u003e] sched_destroy_user+0xd/0xf\n\u003e [\u003cffffffff8023e8c1\u003e] free_uid+0x8a/0xab\n\u003e [\u003cffffffff80233e24\u003e] __put_task_struct+0x3f/0xd3\n\u003e [\u003cffffffff80236708\u003e] delayed_put_task_struct+0x23/0x25\n\u003e [\u003cffffffff8026fda7\u003e] __rcu_process_callbacks+0x8d/0x215\n\u003e [\u003cffffffff8026ff52\u003e] rcu_process_callbacks+0x23/0x44\n\u003e [\u003cffffffff8023a2ae\u003e] __do_softirq+0x79/0xf8\n\u003e [\u003cffffffff8020f8c3\u003e] ? profile_pc+0x2a/0x67\n\u003e [\u003cffffffff8020d38c\u003e] call_softirq+0x1c/0x30\n\u003e [\u003cffffffff8020f689\u003e] do_softirq+0x61/0x9c\n\u003e [\u003cffffffff8023a233\u003e] irq_exit+0x51/0x53\n\u003e [\u003cffffffff8021bd1a\u003e] smp_apic_timer_interrupt+0x77/0xad\n\u003e [\u003cffffffff8020ce3b\u003e] apic_timer_interrupt+0x6b/0x70\n\u003e \u003cEOI\u003e [\u003cffffffff8020b0dd\u003e] ? default_idle+0x43/0x76\n\u003e [\u003cffffffff8020b0db\u003e] ? default_idle+0x41/0x76\n\u003e [\u003cffffffff8020b09a\u003e] ? default_idle+0x0/0x76\n\u003e [\u003cffffffff8020b186\u003e] ? cpu_idle+0x76/0x98\n\nseparate the tg-\u003eshares protection from the task_group lock.\n\nReported-by: Denis V. Lunev \u003cden@openvz.org\u003e\nTested-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "7ad5b3a505e68cfdc342933d6e0fc0eaa5e0a4f7", "tree": "6715ffd8df509d3d53dea581bb97418a21bc7cbc", "parents": [ "fc9b52cd8f5f459b88adcf67c47668425ae31a78" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Fri Feb 08 04:19:53 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:31 2008 -0800" }, "message": "kernel: remove fastcall in kernel/*\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nAcked-by: 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": "75659ca0c10992dcb39258518368a0f6f56e935d", "tree": "5d014ceb2f10158061a23d0d976f9a613d85e659", "parents": [ "fbdde7bd274d74729954190f99afcb1e3d9bbfba", "2dfe485a2c8afa54cb069fcf48476f6c90ea3fdf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 01 11:45:47 2008 +1100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 01 11:45:47 2008 +1100" }, "message": "Merge branch \u0027task_killable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/willy/misc\n\n* \u0027task_killable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/willy/misc: (22 commits)\n Remove commented-out code copied from NFS\n NFS: Switch from intr mount option to TASK_KILLABLE\n Add wait_for_completion_killable\n Add wait_event_killable\n Add schedule_timeout_killable\n Use mutex_lock_killable in vfs_readdir\n Add mutex_lock_killable\n Use lock_page_killable\n Add lock_page_killable\n Add fatal_signal_pending\n Add TASK_WAKEKILL\n exit: Use task_is_*\n signal: Use task_is_*\n sched: Use task_contributes_to_load, TASK_ALL and TASK_NORMAL\n ptrace: Use task_is_*\n power: Use task_is_*\n wait: Use TASK_NORMAL\n proc/base.c: Use task_is_*\n proc/array.c: Use TASK_REPORT\n perfmon: Use task_is_*\n ...\n\nFixed up conflicts in NFS/sunrpc manually..\n" }, { "commit": "5aff0531ee2403b319e89bd04055fc41173e95fd", "tree": "522e811c815d296497cac1e12127fdbfaa4a8da6", "parents": [ "ef9884e6f29bbe1075204f962a00f7533bf7e8f3" ], "author": { "name": "Gerald Stralko", "email": "gerb.stralko@gmail.com", "time": "Thu Jan 31 22:45:23 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Jan 31 22:45:23 2008 +0100" }, "message": "sched: remove unused params\n\nThis removes the extra struct task_struct *p parameter in inc_nr_running\nand dec_nr_running functions.\n\nSigned-off by: Jerry Stralko \u003cgerb.stralko@gmail.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "95c354fe9f7d6decc08a92aa26eb233ecc2155bf", "tree": "ec9267032ea875e84216cfb20acb2cfc7c62149f", "parents": [ "a95d67f87e1a5f1b4429be3ba3bf7b4051657908" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Wed Jan 30 13:31:20 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Jan 30 13:31:20 2008 +0100" }, "message": "spinlock: lockbreak cleanup\n\nThe break_lock data structure and code for spinlocks is quite nasty.\nNot only does it double the size of a spinlock but it changes locking to\na potentially less optimal trylock.\n\nPut all of that under CONFIG_GENERIC_LOCKBREAK, and introduce a\n__raw_spin_is_contended that uses the lock data itself to determine whether\nthere are waiters on the lock, to be used if CONFIG_GENERIC_LOCKBREAK is\nnot set.\n\nRename need_lockbreak to spin_needbreak, make it use spin_is_contended to\ndecouple it from the spinlock implementation, and make it typesafe (rwlocks\ndo not have any need_lockbreak sites -- why do they even get bloated up\nwith that break_lock then?).\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "5fb5e6de55860a99c2d8fe7e0c8222d5c53d8464", "tree": "7a9935bfa8b20c56ddc865c130f17d2f42444ce8", "parents": [ "19ef9309273d26cb005cb23e6a370353dca91099" ], "author": { "name": "Nick Piggin", "email": "nickpiggin@yahoo.com.au", "time": "Fri Jan 25 21:08:34 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Jan 25 21:08:34 2008 +0100" }, "message": "sched: print backtrace of running tasks too\n\nThe attached patch is something really simple that can sometimes help\nin getting more info out of a hung system.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" } ], "next": "cc203d2422004498909c4886d1b94a2e388d973e" }