- 03 Oct, 2014 1 commit
-
-
Dave Hansen authored
Commit: cebf15eb ("x86, sched: Add new topology for multi-NUMA-node CPUs") some code to try to detect the situation where we have a NUMA node inside of the "DIE" sched domain. It detected this by looking for cpus which match_die() but do not match NUMA nodes via topology_same_node(). I wrote it up as: if (match_die(c, o) == !topology_same_node(c, o)) which actually seemed to work some of the time, albiet accidentally. It should have been doing an &&, not an ==. This code essentially chopped off the "DIE" domain on one of Andrew Morton's systems. He reported that this patch fixed his issue. Signed-off-by:
Dave Hansen <dave.hansen@linux.intel.com> Reported-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Dave Hansen <dave@sr71.net> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: David Rientjes <rientjes@google.com> Cc: Igor Mammedov <imammedo@redhat.com> Cc: Jan Kiszka <jan.kiszka@siemens.com> Cc: Lan Tianyu <tianyu.lan@intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Toshi Kani <toshi.kani@hp.com> Link: http://lkml.kernel.org/r/20140930214546.FD481CFF@viggo.jf.intel.comSigned-off-by:
Ingo Molnar <mingo@kernel.org>
-
- 24 Sep, 2014 15 commits
-
-
Dave Hansen authored
I'm getting the spew below when booting with Haswell (Xeon E5-2699 v3) CPUs and the "Cluster-on-Die" (CoD) feature enabled in the BIOS. It seems similar to the issue that some folks from AMD ran in to on their systems and addressed in this commit: 161270fc ("x86/smp: Fix topology checks on AMD MCM CPUs") Both these Intel and AMD systems break an assumption which is being enforced by topology_sane(): a socket may not contain more than one NUMA node. AMD special-cased their system by looking for a cpuid flag. The Intel mode is dependent on BIOS options and I do not know of a way which it is enumerated other than the tables being parsed during the CPU bringup process. In other words, we have to trust the ACPI tables <shudder>. This detects the situation where a NUMA node occurs at a place in the middle of the "CPU" sched domains. It replaces the default topology with one that relies on the NUMA information from the firmware (SRAT table) for all levels of sched domains above the hyperthreads. This also fixes a sysfs bug. We used to freak out when we saw the "mc" group cross a node boundary, so we stopped building the MC group. MC gets exported as the 'core_siblings_list' in /sys/devices/system/cpu/cpu*/topology/ and this caused CPUs with the same 'physical_package_id' to not be listed together in 'core_siblings_list'. This violates a statement from Documentation/ABI/testing/sysfs-devices-system-cpu: core_siblings: internal kernel map of cpu#'s hardware threads within the same physical_package_id. core_siblings_list: human-readable list of the logical CPU numbers within the same physical_package_id as cpu#. The sysfs effects here cause an issue with the hwloc tool where it gets confused and thinks there are more sockets than are physically present. Before this patch, there are two packages: # cd /sys/devices/system/cpu/ # cat cpu*/topology/physical_package_id | sort | uniq -c 18 0 18 1 But 4 _sets_ of core siblings: # cat cpu*/topology/core_siblings_list | sort | uniq -c 9 0-8 9 18-26 9 27-35 9 9-17 After this set, there are only 2 sets of core siblings, which is what we expect for a 2-socket system. # cat cpu*/topology/physical_package_id | sort | uniq -c 18 0 18 1 # cat cpu*/topology/core_siblings_list | sort | uniq -c 18 0-17 18 18-35 Example spew: ... NMI watchdog: enabled on all CPUs, permanently consumes one hw-PMU counter. #2 #3 #4 #5 #6 #7 #8 .... node #1, CPUs: #9 ------------[ cut here ]------------ WARNING: CPU: 9 PID: 0 at /home/ak/hle/linux-hle-2.6/arch/x86/kernel/smpboot.c:306 topology_sane.isra.2+0x74/0x90() sched: CPU #9's mc-sibling CPU #0 is not on the same node! [node: 1 != 0]. Ignoring dependency. Modules linked in: CPU: 9 PID: 0 Comm: swapper/9 Not tainted 3.17.0-rc1-00293-g8e01c4d-dirty #631 Hardware name: Intel Corporation S2600WTT/S2600WTT, BIOS GRNDSDP1.86B.0036.R05.1407140519 07/14/2014 0000000000000009 ffff88046ddabe00 ffffffff8172e485 ffff88046ddabe48 ffff88046ddabe38 ffffffff8109691d 000000000000b001 0000000000000009 ffff88086fc12580 000000000000b020 0000000000000009 ffff88046ddabe98 Call Trace: [<ffffffff8172e485>] dump_stack+0x45/0x56 [<ffffffff8109691d>] warn_slowpath_common+0x7d/0xa0 [<ffffffff8109698c>] warn_slowpath_fmt+0x4c/0x50 [<ffffffff81074f94>] topology_sane.isra.2+0x74/0x90 [<ffffffff8107530e>] set_cpu_sibling_map+0x31e/0x4f0 [<ffffffff8107568d>] start_secondary+0x1ad/0x240 ---[ end trace 3fe5f587a9fcde61 ]--- #10 #11 #12 #13 #14 #15 #16 #17 .... node #2, CPUs: #18 #19 #20 #21 #22 #23 #24 #25 #26 .... node #3, CPUs: #27 #28 #29 #30 #31 #32 #33 #34 #35 Signed-off-by:
-
Kirill Tkhai authored
Some time ago PREEMPT_NEED_RESCHED was implemented, so reschedule technics is a little more difficult now. Signed-off-by:
Kirill Tkhai <ktkhai@parallels.com> Signed-off-by:
-
Kirill Tkhai authored
Probability of use-after-free isn't zero in this place. Signed-off-by:
-
Kirill Tkhai authored
Nothing is locked there, so label's name only confuses a reader. Signed-off-by:
-
Kirill Tkhai authored
dl_bw_of() dereferences rq->rd which has to have RCU read lock held. Probability of use-after-free isn't zero here. Also add lockdep assert into dl_bw_cpus(). Signed-off-by:
-
Kirill Tkhai authored
Combine two branches which do the same. Signed-off-by:
-
Peter Zijlstra authored
Kirill found that there's a subtle race in the __ARCH_WANT_UNLOCKED_CTXSW code, and instead of fixing it, remove the entire exception because neither arch that uses it seems to actually still require it. Boot tested on mips64el (qemu) only. Signed-off-by:
Kirill Tkhai <tkhai@yandex.ru> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Davidlohr Bueso <davidlohr@hp.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: James Hogan <james.hogan@imgtec.com> Cc: Kees Cook <keescook@chromium.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Paul Burton <paul.burton@imgtec.com> Cc: Qais Yousef <qais.yousef@imgtec.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Tony Luck <tony.luck@intel.com> Cc: oleg@redhat.com Cc: linux@roeck-us.net Cc: linux-ia64@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Link: http://lkml.kernel.org/r/20140923150641.GH3312@worktop.programming.kicks-ass.netSigned-off-by:
-
Oleg Nesterov authored
read_lock_irqsave(tasklist_lock) in print_rq() looks strange. We do not need to disable irqs, and they are already disabled by the caller. And afaics this lock buys nothing, we can rely on rcu_read_lock(). In this case it makes sense to also move rcu_read_lock/unlock from the caller to print_rq(). Signed-off-by:
Oleg Nesterov <oleg@redhat.com> Signed-off-by:
-
Oleg Nesterov authored
1. read_lock(tasklist_lock) does not need to disable irqs. 2. ->mm != NULL is a common mistake, use PF_KTHREAD. 3. The second ->mm check can be simply removed. 4. task_rq_lock() looks better than raw_spin_lock(&p->pi_lock) + __task_rq_lock(). Signed-off-by:
-
Oleg Nesterov authored
tg_has_rt_tasks() wants to find an RT task in this task_group, but task_rq(p)->rt.tg wrongly checks the root rt_rq. Signed-off-by:
-
Nicolas Pitre authored
The code in find_idlest_cpu() looks for the CPU with the smallest load. However, if multiple CPUs are idle, the first idle CPU is selected irrespective of the depth of its idle state. Among the idle CPUs we should pick the one with with the shallowest idle state, or the latest to have gone idle if all idle CPUs are in the same state. The later applies even when cpuidle is configured out. This patch doesn't cover the following issues: - The idle exit latency of a CPU might be larger than the time needed to migrate the waking task to an already running CPU with sufficient capacity, and therefore performance would benefit from task packing in such case (in most cases task packing is about power saving). - Some idle states have a non negligible and non abortable entry latency which needs to run to completion before the exit latency can start. A concurrent patch series is making this info available to the cpuidle core. Once available, the entry latency with the idle timestamp could determine when the exit latency may be effective. Those issues will be handled in due course. In the mean time, what is implemented here should improve things already compared to the current state of affairs. Based on an initial patch from Daniel Lezcano. Signed-off-by:
Nicolas Pitre <nico@linaro.org> Signed-off-by:
-
Daniel Lezcano authored
When the cpu enters idle, it stores the cpuidle state pointer in its struct rq instance which in turn could be used to make a better decision when balancing tasks. As soon as the cpu exits its idle state, the struct rq reference is cleared. There are a couple of situations where the idle state pointer could be changed while it is being consulted: 1. For x86/acpi with dynamic c-states, when a laptop switches from battery to AC that could result on removing the deeper idle state. The acpi driver triggers: 'acpi_processor_cst_has_changed' 'cpuidle_pause_and_lock' 'cpuidle_uninstall_idle_handler' 'kick_all_cpus_sync'. All cpus will exit their idle state and the pointed object will be set to NULL. 2. The cpuidle driver is unloaded. Logically that could happen but not in practice because the drivers are always compiled in and 95% of them are not coded to unregister themselves. In any case, the unloading code must call 'cpuidle_unregister_device', that calls 'cpuidle_pause_and_lock' leading to 'kick_all_cpus_sync' as mentioned above. A race can happen if we use the pointer and then one of these two scenarios occurs at the same moment. In order to be safe, the idle state pointer stored in the rq must be used inside a rcu_read_lock section where we are protected with the 'rcu_barrier' in the 'cpuidle_uninstall_idle_handler' function. The idle_get_state() and idle_put_state() accessors should be used to that effect. Signed-off-by:
Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by:
-
Juri Lelli authored
Users can perform clustered scheduling using the cpuset facility. After an exclusive cpuset is created, task migrations happen only between CPUs belonging to the same cpuset. Inter- cpuset migrations can only happen when the user requires so, moving a task between different cpusets. This behaviour is broken in SCHED_DEADLINE, as currently spurious inter- cpuset migration may happen without user intervention. This patch fix the problem (and shuffles the code a bit to improve clarity). Signed-off-by:
Juri Lelli <juri.lelli@arm.com> Signed-off-by:
-
Juri Lelli authored
When a task is using SCHED_DEADLINE and the user setschedules it to a different class its sched_dl_entity static parameters are not cleaned up. This causes a bug if the user sets it back to SCHED_DEADLINE with the same parameters again. The problem resides in the check we perform at the very beginning of dl_overflow(): if (new_bw == p->dl.dl_bw) return 0; This condition is met in the case depicted above, so the function returns and dl_b->total_bw is not updated (the p->dl.dl_bw is not added to it). After this, admission control is broken. This patch fixes the thing, properly clearing static parameters for a task that ceases to use SCHED_DEADLINE. Reported-by:
Daniele Alessandrelli <daniele.alessandrelli@gmail.com> Reported-by:
Daniel Wagner <daniel.wagner@bmw-carit.de> Reported-by:
Vincent Legout <vincent@legout.info> Tested-by:
Luca Abeni <luca.abeni@unitn.it> Tested-by:
-
Oleg Nesterov authored
current->state == TASK_DEAD means that the task is doing its last schedule(), page fault is obviously impossible at this stage. Signed-off-by:
Mel Gorman <mgorman@suse.de> Acked-by:
Rik van Riel <riel@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/r/20140921194743.GA30114@redhat.comSigned-off-by:
-
- 21 Sep, 2014 1 commit
-
-
Zhihui Zhang authored
Signed-off-by:
Zhihui Zhang <zzhsuny@gmail.com> Cc: peterz@infradead.org Link: http://lkml.kernel.org/r/1411262676-19928-1-git-send-email-zzhsuny@gmail.comSigned-off-by:
-
- 19 Sep, 2014 21 commits
-
-
Vincent Guittot authored
Currently the task always wakes affine on this_cpu if the latter is idle. Before waking up the task on this_cpu, we check that this_cpu capacity is not significantly reduced because of RT tasks or irq activity. Use case where the number of irq and/or the time spent under irq is important will take benefit of this because the task that is woken up by irq or softirq will not use the same CPU than irq (and softirq) but a idle one. Signed-off-by:
Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by:
-
Vincent Guittot authored
Use the new arch_scale_cpu_capacity() scheduler facility in order to reflect the original capacity of a CPU instead of arch_scale_freq_capacity() which is more linked to a scaling of the capacity linked to the frequency. Signed-off-by:
-
Vincent Guittot authored
'capacity_orig' is only changed for systems with an SMT sched_domain level in order to reflect the lower capacity of CPUs. Heterogenous systems also have to reflect an original capacity that is different from the default value. Create a more generic function arch_scale_cpu_capacity that can be also used by non SMT platforms to set capacity_orig. The __weak implementation of arch_scale_cpu_capacity() is the previous SMT variant, in order to keep backward compatibility with the use of capacity_orig. arch_scale_smt_capacity() and default_scale_smt_capacity() have been removed as they were not used elsewhere than in arch_scale_cpu_capacity(). Signed-off-by:
Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Reviewed-by:
Preeti U. Murthy <preeti@linux.vnet.ibm.com> [ Added default_scale_cpu_capacity() back. ] Signed-off-by:
-
Vincent Guittot authored
The computation of avg_load and avg_load_per_task should only take into account the number of CFS tasks. The non-CFS tasks are already taken into account by decreasing the CPU's capacity and they will be tracked in the CPU's utilization (group_utilization) of the next patches. Reviewed-by:
-
Vincent Guittot authored
In wake_affine() I have tried to understand the meaning of the condition: (this_load <= load && this_load + target_load(prev_cpu, idx) <= tl_per_task) but I failed to find a use case that can take advantage of it and I haven't found clear description in the previous commit's log. Futhermore, the comment of the condition refers to the task_hot function that was used before being replaced by the current condition: /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and * there is no bad imbalance. */ If we look more deeply the below condition: this_load + target_load(prev_cpu, idx) <= tl_per_task When sync is clear, we have: tl_per_task = runnable_load_avg / nr_running this_load = max(runnable_load_avg, cpuload[idx]) target_load = max(runnable_load_avg', cpuload'[idx]) It implies that runnable_load_avg == 0 and nr_running <= 1 in order to match the condition. This implies that runnable_load_avg == 0 too because of the condition: this_load <= load. but if this _load is null, 'balanced' is already set and the test is redundant. If sync is set, it's not as straight forward as above (especially if cgroup are involved) but the policy should be similar as we have removed a task that's going to sleep in order to get a more accurate load and this_load values. The current conclusion is that these additional condition don't give any benefit so we can remove them. Signed-off-by:
-
Vincent Guittot authored
The imbalance flag can stay set whereas there is no imbalance. Let assume that we have 3 tasks that run on a dual cores /dual cluster system. We will have some idle load balance which are triggered during tick. Unfortunately, the tick is also used to queue background work so we can reach the situation where short work has been queued on a CPU which already runs a task. The load balance will detect this imbalance (2 tasks on 1 CPU and an idle CPU) and will try to pull the waiting task on the idle CPU. The waiting task is a worker thread that is pinned on a CPU so an imbalance due to pinned task is detected and the imbalance flag is set. Then, we will not be able to clear the flag because we have at most 1 task on each CPU but the imbalance flag will trig to useless active load balance between the idle CPU and the busy CPU. We need to reset of the imbalance flag as soon as we have reached a balanced state. If all tasks are pinned, we don't consider that as a balanced state and let the imbalance flag set. Signed-off-by:
-
Aaron Tomlin authored
Currently in the event of a stack overrun a call to schedule() does not check for this type of corruption. This corruption is often silent and can go unnoticed. However once the corrupted region is examined at a later stage, the outcome is undefined and often results in a sporadic page fault which cannot be handled. This patch checks for a stack overrun and takes appropriate action since the damage is already done, there is no point in continuing. Signed-off-by:
Aaron Tomlin <atomlin@redhat.com> Signed-off-by:
-
Aaron Tomlin authored
This facility is used in a few places so let's introduce a helper function to improve code readability. Signed-off-by:
-
Aaron Tomlin authored
Tasks get their end of stack set to STACK_END_MAGIC with the aim to catch stack overruns. Currently this feature does not apply to init_task. This patch removes this restriction. Note that a similar patch was posted by Prarit Bhargava some time ago but was never merged: http://marc.info/?l=linux-kernel&m=127144305403241&w=2Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Cc: aneesh.kumar@linux.vnet.ibm.com Cc: dzickus@redhat.com Cc: bmr@redhat.com Cc: jcastillo@redhat.com Cc: jgh@redhat.com Cc: minchan@kernel.org Cc: tglx@linutronix.de Cc: hannes@cmpxchg.org Cc: Alex Thorlton <athorlton@sgi.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daeseok Youn <daeseok.youn@gmail.com> Cc: David Rientjes <rientjes@google.com> Cc: Fabian Frederick <fabf@skynet.be> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Kees Cook <keescook@chromium.org> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Michael Opdenacker <michael.opdenacker@free-electrons.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Seiji Aguchi <seiji.aguchi@hds.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Vladimir Davydov <vdavydov@parallels.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1410527779-8133-2-git-send-email-atomlin@redhat.comSigned-off-by:
-
Kirill Tkhai authored
If a task is queued but not running on it rq, we can simply migrate it without migration thread and switching of context. Signed-off-by:
-
Kirill Tkhai authored
1) Nobody calls pick_dl_task() with negative cpu, it's old RT leftover. 2) If p->nr_cpus_allowed is 1, than the affinity has just been changed in set_cpus_allowed_ptr(); we'll pick it just earlier than migration thread. Signed-off-by:
-
Kirill Tkhai authored
_pick_next_task_rt() never returns NULL. Signed-off-by:
-
Kirill Tkhai authored
Signed-off-by:
-
Kirill Tkhai authored
new_cpu is reassigned below, so we do not need this here. Signed-off-by:
-
Kirill Tkhai authored
schedule(), io_schedule() and schedule_timeout() always return with TASK_RUNNING state set, so one more setting is unnecessary. (All places in patch are visible good, only exception is kiblnd_scheduler() from: drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c Its schedule() is one line above standard 3 lines of unified diff) No places where set_current_state() is used for mb(). Signed-off-by: -
Rik van Riel authored
The sig->stats_lock nests inside the tasklist_lock and the sighand->siglock in __exit_signal and wait_task_zombie. However, both of those locks can be taken from irq context, which means we need to use the interrupt safe variant of read_seqbegin_or_lock. This blocks interrupts when the "lock" branch is taken (seq is odd), preventing the lock inversion. On the first (lockless) pass through the loop, irqs are not blocked. Reported-by:
Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by:
-
Rik van Riel authored
There are cases where read_seqbegin_or_lock() needs to block irqs, because the seqlock in question nests inside a lock that is also be taken from irq context. Add read_seqbegin_or_lock_irqsave() and done_seqretry_irqrestore(), which are almost identical to read_seqbegin_or_lock() and done_seqretry(). Signed-off-by:
-
Chuansheng Liu authored
Currently kick_all_cpus_sync() or smp_call_function() can not break the polling idle cpu immediately. Instead using wake_up_all_idle_cpus() which can wake up the polling idle cpu quickly is much more helpful for power. Signed-off-by:
Chuansheng Liu <chuansheng.liu@intel.com> Signed-off-by:
-
Chuansheng Liu authored
Currently kick_all_cpus_sync() can break non-polling idle cpus thru IPI interrupts. But sometimes we need to break the polling idle cpus immediately to reselect the suitable c-state, also for non-idle cpus, we need to do nothing if we try to wake up them. Here adding one new function wake_up_all_idle_cpus() to let all cpus out of idle based on function wake_up_if_idle(). Signed-off-by:
-
Chuansheng Liu authored
Implementing one new API wake_up_if_idle(), which is used to wake up the idle CPU. Suggested-by:
Andy Lutomirski <luto@amacapital.net> Signed-off-by:
-
Rik van Riel authored
The code in task_numa_compare() will only examine at most one idle CPU per node, because they all have the same score. However, some idle CPUs are better candidates than others, due to busy or idle SMT siblings, etc... The scheduler has logic to find the best CPU within an LLC to place a task. The NUMA code should probably use it. This seems to reduce the standard deviation for single instance SPECjbb2005 with a low warehouse count on my 4 node test system. Signed-off-by:
-
- 16 Sep, 2014 2 commits
-
-
Juri Lelli authored
Add an appendix providing a simple self-contained code snippet showing how SCHED_DEADLINE reservations can be created by application developers. Signed-off-by:
Henrik Austad <henrik@austad.us> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Dario Faggioli <raistlin@linux.it> Cc: Juri Lelli <juri.lelli@gmail.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/r/1410256636-26171-6-git-send-email-juri.lelli@arm.com [ Fixed some whitespace inconsistencies. ] Signed-off-by:
-
Juri Lelli authored
Add an appendix briefly describing tools that can be used to test SCHED_DEADLINE (and the scheduler in general). Links to where source code of the tools is hosted are also provided. Signed-off-by:
-
