| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Just like the ttwu_queue_remote() IPI, make use of _TIF_POLLING_NRFLAG
to avoid sending IPIs to idle CPUs.
[ mingo: Fix UP build bug. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20200526161907.953304789@infradead.org
|
|
The recent commit: 90b5363acd47 ("sched: Clean up scheduler_ipi()")
got smp_call_function_single_async() subtly wrong. Even though it will
return -EBUSY when trying to re-use a csd, that condition is not
atomic and still requires external serialization.
The change in kick_ilb() got this wrong.
While on first reading kick_ilb() has an atomic test-and-set that
appears to serialize the use, the matching 'release' is not in the
right place to actually guarantee this serialization.
Rework the nohz_idle_balance() trigger so that the release is in the
IPI callback and thus guarantees the required serialization for the
CSD.
Fixes: 90b5363acd47 ("sched: Clean up scheduler_ipi()")
Reported-by: Qian Cai <cai@lca.pw>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Cc: mgorman@techsingularity.net
Link: https://lore.kernel.org/r/20200526161907.778543557@infradead.org
|
|
We are going to rely on the loosening of RCU callback semantics,
introduced by this commit:
806f04e9fd2c: ("rcu: Allow for smp_call_function() running callbacks from idle")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
The previous commit:
c6e7bd7afaeb: ("sched/core: Optimize ttwu() spinning on p->on_cpu")
avoids spinning on p->on_rq when the task is descheduling, but only if the
wakee is on a CPU that does not share cache with the waker.
This patch offloads the activation of the wakee to the CPU that is about to
go idle if the task is the only one on the runqueue. This potentially allows
the waker task to continue making progress when the wakeup is not strictly
synchronous.
This is very obvious with netperf UDP_STREAM running on localhost. The
waker is sending packets as quickly as possible without waiting for any
reply. It frequently wakes the server for the processing of packets and
when netserver is using local memory, it quickly completes the processing
and goes back to idle. The waker often observes that netserver is on_rq
and spins excessively leading to a drop in throughput.
This is a comparison of 5.7-rc6 against "sched: Optimize ttwu() spinning
on p->on_cpu" and against this patch labeled vanilla, optttwu-v1r1 and
localwakelist-v1r2 respectively.
5.7.0-rc6 5.7.0-rc6 5.7.0-rc6
vanilla optttwu-v1r1 localwakelist-v1r2
Hmean send-64 251.49 ( 0.00%) 258.05 * 2.61%* 305.59 * 21.51%*
Hmean send-128 497.86 ( 0.00%) 519.89 * 4.43%* 600.25 * 20.57%*
Hmean send-256 944.90 ( 0.00%) 997.45 * 5.56%* 1140.19 * 20.67%*
Hmean send-1024 3779.03 ( 0.00%) 3859.18 * 2.12%* 4518.19 * 19.56%*
Hmean send-2048 7030.81 ( 0.00%) 7315.99 * 4.06%* 8683.01 * 23.50%*
Hmean send-3312 10847.44 ( 0.00%) 11149.43 * 2.78%* 12896.71 * 18.89%*
Hmean send-4096 13436.19 ( 0.00%) 13614.09 ( 1.32%) 15041.09 * 11.94%*
Hmean send-8192 22624.49 ( 0.00%) 23265.32 * 2.83%* 24534.96 * 8.44%*
Hmean send-16384 34441.87 ( 0.00%) 36457.15 * 5.85%* 35986.21 * 4.48%*
Note that this benefit is not universal to all wakeups, it only applies
to the case where the waker often spins on p->on_rq.
The impact can be seen from a "perf sched latency" report generated from
a single iteration of one packet size:
-----------------------------------------------------------------------------------------------------------------
Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at |
-----------------------------------------------------------------------------------------------------------------
vanilla
netperf:4337 | 21709.193 ms | 2932 | avg: 0.002 ms | max: 0.041 ms | max at: 112.154512 s
netserver:4338 | 14629.459 ms | 5146990 | avg: 0.001 ms | max: 1615.864 ms | max at: 140.134496 s
localwakelist-v1r2
netperf:4339 | 29789.717 ms | 2460 | avg: 0.002 ms | max: 0.059 ms | max at: 138.205389 s
netserver:4340 | 18858.767 ms | 7279005 | avg: 0.001 ms | max: 0.362 ms | max at: 135.709683 s
-----------------------------------------------------------------------------------------------------------------
Note that the average wakeup delay is quite small on both the vanilla
kernel and with the two patches applied. However, there are significant
outliers with the vanilla kernel with the maximum one measured as 1615
milliseconds with a vanilla kernel but never worse than 0.362 ms with
both patches applied and a much higher rate of context switching.
Similarly a separate profile of cycles showed that 2.83% of all cycles
were spent in try_to_wake_up() with almost half of the cycles spent
on spinning on p->on_rq. With the two patches, the percentage of cycles
spent in try_to_wake_up() drops to 1.13%
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Jirka Hladky <jhladky@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: valentin.schneider@arm.com
Cc: Hillf Danton <hdanton@sina.com>
Cc: Rik van Riel <riel@surriel.com>
Link: https://lore.kernel.org/r/20200524202956.27665-3-mgorman@techsingularity.net
|
|
Both Rik and Mel reported seeing ttwu() spend significant time on:
smp_cond_load_acquire(&p->on_cpu, !VAL);
Attempt to avoid this by queueing the wakeup on the CPU that owns the
p->on_cpu value. This will then allow the ttwu() to complete without
further waiting.
Since we run schedule() with interrupts disabled, the IPI is
guaranteed to happen after p->on_cpu is cleared, this is what makes it
safe to queue early.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Jirka Hladky <jhladky@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: valentin.schneider@arm.com
Cc: Hillf Danton <hdanton@sina.com>
Cc: Rik van Riel <riel@surriel.com>
Link: https://lore.kernel.org/r/20200524202956.27665-2-mgorman@techsingularity.net
|
|
When users write some huge number into cpu.cfs_quota_us or
cpu.rt_runtime_us, overflow might happen during to_ratio() shifts of
schedulable checks.
to_ratio() could be altered to avoid unnecessary internal overflow, but
min_cfs_quota_period is less than 1 << BW_SHIFT, so a cutoff would still
be needed. Set a cap MAX_BW for cfs_quota_us and rt_runtime_us to
prevent overflow.
Signed-off-by: Huaixin Chang <changhuaixin@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Link: https://lkml.kernel.org/r/20200425105248.60093-1-changhuaixin@linux.alibaba.com
|
|
Get the noinstr section and annotation markers to base the RCU parts on.
|
|
There is nothing architecture-specific about scs_overflow_check() as
it's just a trivial wrapper around scs_corrupted().
For parity with task_stack_end_corrupted(), rename scs_corrupted() to
task_scs_end_corrupted() and call it from schedule_debug() when
CONFIG_SCHED_STACK_END_CHECK_is enabled, which better reflects its
purpose as a debug feature to catch inadvertent overflow of the SCS.
Finally, remove the unused scs_overflow_check() function entirely.
This has absolutely no impact on architectures that do not support SCS
(currently arm64 only).
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
|
|
This change adds generic support for Clang's Shadow Call Stack,
which uses a shadow stack to protect return addresses from being
overwritten by an attacker. Details are available here:
https://clang.llvm.org/docs/ShadowCallStack.html
Note that security guarantees in the kernel differ from the ones
documented for user space. The kernel must store addresses of
shadow stacks in memory, which means an attacker capable reading
and writing arbitrary memory may be able to locate them and hijack
control flow by modifying the stacks.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>
[will: Numerous cosmetic changes]
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Now that the scheduler IPI is trivial and simple again there is no point to
have the little function out of line. This simplifies the effort of
constraining the instrumentation nicely.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200505134058.453581595@linutronix.de
|
|
The scheduler IPI has grown weird and wonderful over the years, time
for spring cleaning.
Move all the non-trivial stuff out of it and into a regular smp function
call IPI. This then reduces the schedule_ipi() to most of it's former NOP
glory and ensures to keep the interrupt vector lean and mean.
Aside of that avoiding the full irq_enter() in the x86 IPI implementation
is incorrect as scheduler_ipi() can be instrumented. To work around that
scheduler_ipi() had an irq_enter/exit() hack when heavy work was
pending. This is gone now.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Link: https://lkml.kernel.org/r/20200505134058.361859938@linutronix.de
|
|
Currently root_task_group.shares and cfs_bandwidth are initialized for
each online cpu, which not necessary.
Let's take it out to do it only once.
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200423214443.29994-1-richard.weiyang@gmail.com
|
|
In the CPU-offline process, it calls mmdrop() after idle entry and the
subsequent call to cpuhp_report_idle_dead(). Once execution passes the
call to rcu_report_dead(), RCU is ignoring the CPU, which results in
lockdep complaining when mmdrop() uses RCU from either memcg or
debugobjects below.
Fix it by cleaning up the active_mm state from BP instead. Every arch
which has CONFIG_HOTPLUG_CPU should have already called idle_task_exit()
from AP. The only exception is parisc because it switches them to
&init_mm unconditionally (see smp_boot_one_cpu() and smp_cpu_init()),
but the patch will still work there because it calls mmgrab(&init_mm) in
smp_cpu_init() and then should call mmdrop(&init_mm) in finish_cpu().
WARNING: suspicious RCU usage
-----------------------------
kernel/workqueue.c:710 RCU or wq_pool_mutex should be held!
other info that might help us debug this:
RCU used illegally from offline CPU!
Call Trace:
dump_stack+0xf4/0x164 (unreliable)
lockdep_rcu_suspicious+0x140/0x164
get_work_pool+0x110/0x150
__queue_work+0x1bc/0xca0
queue_work_on+0x114/0x120
css_release+0x9c/0xc0
percpu_ref_put_many+0x204/0x230
free_pcp_prepare+0x264/0x570
free_unref_page+0x38/0xf0
__mmdrop+0x21c/0x2c0
idle_task_exit+0x170/0x1b0
pnv_smp_cpu_kill_self+0x38/0x2e0
cpu_die+0x48/0x64
arch_cpu_idle_dead+0x30/0x50
do_idle+0x2f4/0x470
cpu_startup_entry+0x38/0x40
start_secondary+0x7a8/0xa80
start_secondary_resume+0x10/0x14
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Link: https://lkml.kernel.org/r/20200401214033.8448-1-cai@lca.pw
|
|
Introduce a new function put_prev_task_balance() to do the balance
when necessary, and then put previous task back to the run queue.
This function is extracted from pick_next_task() to prepare for
future usage by other type of task picking logic.
No functional change.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lkml.kernel.org/r/5a99860cf66293db58a397d6248bcb2eee326776.1587464698.git.yu.c.chen@intel.com
|
|
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull in Christoph Hellwig's series that changes the sysctl's ->proc_handler
methods to take kernel pointers instead. It gets rid of the set_fs address
space overrides used by BPF. As per discussion, pull in the feature branch
into bpf-next as it relates to BPF sysctl progs.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200427071508.GV23230@ZenIV.linux.org.uk/T/
|
|
A running task's state can be sampled in a consistent manner (for example,
for diagnostic purposes) simply by invoking smp_call_function_single()
on its CPU, which may be obtained using task_cpu(), then having the
IPI handler verify that the desired task is in fact still running.
However, if the task is not running, this sampling can in theory be done
immediately and directly. In practice, the task might start running at
any time, including during the sampling period. Gaining a consistent
sample of a not-running task therefore requires that something be done
to lock down the target task's state.
This commit therefore adds a try_invoke_on_locked_down_task() function
that invokes a specified function if the specified task can be locked
down, returning true if successful and if the specified function returns
true. Otherwise this function simply returns false. Given that the
function passed to try_invoke_on_nonrunning_task() might be invoked with
a runqueue lock held, that function had better be quite lightweight.
The function is passed the target task's task_struct pointer and the
argument passed to try_invoke_on_locked_down_task(), allowing easy access
to task state and to a location for further variables to be passed in
and out.
Note that the specified function will be called even if the specified
task is currently running. The function can use ->on_rq and task_curr()
to quickly and easily determine the task's state, and can return false
if this state is not to the function's liking. The caller of the
try_invoke_on_locked_down_task() would then see the false return value,
and could take appropriate action, for example, trying again later or
sending an IPI if matters are more urgent.
It is expected that use cases such as the RCU CPU stall warning code will
simply return false if the task is currently running. However, there are
use cases involving nohz_full CPUs where the specified function might
instead fall back to an alternative sampling scheme that relies on heavier
synchronization (such as memory barriers) in the target task.
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Ben Segall <bsegall@google.com>
Cc: Mel Gorman <mgorman@suse.de>
[ paulmck: Apply feedback from Peter Zijlstra and Steven Rostedt. ]
[ paulmck: Invoke if running to handle feedback from Mathieu Desnoyers. ]
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
|
|
Instead of having all the sysctl handlers deal with user pointers, which
is rather hairy in terms of the BPF interaction, copy the input to and
from userspace in common code. This also means that the strings are
always NUL-terminated by the common code, making the API a little bit
safer.
As most handler just pass through the data to one of the common handlers
a lot of the changes are mechnical.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
uclamp_fork() resets the uclamp values to their default when the
reset-on-fork flag is set. It also checks whether the task has a RT
policy, and sets its uclamp.min to 1024 accordingly. However, during
reset-on-fork, the task's policy is lowered to SCHED_NORMAL right after,
hence leading to an erroneous uclamp.min setting for the new task if it
was forked from RT.
Fix this by removing the unnecessary check on rt_task() in
uclamp_fork() as this doesn't make sense if the reset-on-fork flag is
set.
Fixes: 1a00d999971c ("sched/uclamp: Set default clamps for RT tasks")
Reported-by: Chitti Babu Theegala <ctheegal@codeaurora.org>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Patrick Bellasi <patrick.bellasi@matbug.net>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lkml.kernel.org/r/20200416085956.217587-1-qperret@google.com
|
|
The following commit:
5e83eafbfd3b ("sched/fair: Remove the rq->cpu_load[] update code")
eliminated the last use case for rq->last_load_update_tick, so remove
the field as well.
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/1584710495-308969-1-git-send-email-vincent.donnefort@arm.com
|
|
The kernel test robot triggered a warning with the following race:
task-ctx A interrupt-ctx B
worker
-> process_one_work()
-> work_item()
-> schedule();
-> sched_submit_work()
-> wq_worker_sleeping()
-> ->sleeping = 1
atomic_dec_and_test(nr_running)
__schedule(); *interrupt*
async_page_fault()
-> local_irq_enable();
-> schedule();
-> sched_submit_work()
-> wq_worker_sleeping()
-> if (WARN_ON(->sleeping)) return
-> __schedule()
-> sched_update_worker()
-> wq_worker_running()
-> atomic_inc(nr_running);
-> ->sleeping = 0;
-> sched_update_worker()
-> wq_worker_running()
if (!->sleeping) return
In this context the warning is pointless everything is fine.
An interrupt before wq_worker_sleeping() will perform the ->sleeping
assignment (0 -> 1 > 0) twice.
An interrupt after wq_worker_sleeping() will trigger the warning and
nr_running will be decremented (by A) and incremented once (only by B, A
will skip it). This is the case until the ->sleeping is zeroed again in
wq_worker_running().
Remove the WARN statement because this condition may happen. Document
that preemption around wq_worker_sleeping() needs to be disabled to
protect ->sleeping and not just as an optimisation.
Fixes: 6d25be5782e48 ("sched/core, workqueues: Distangle worker accounting from rq lock")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Link: https://lkml.kernel.org/r/20200327074308.GY11705@shao2-debian
|
|
sched/core.c uses update_avg() for rq->avg_idle and sched/fair.c uses an
open-coded version (with the exact same decay factor) for
rq->avg_scan_cost. On top of that, select_idle_cpu() expects to be able to
compare these two fields.
The only difference between the two is that rq->avg_scan_cost is computed
using a pure division rather than a shift. Turns out it actually matters,
first of all because the shifted value can be negative, and the standard
has this to say about it:
"""
The result of E1 >> E2 is E1 right-shifted E2 bit positions. [...] If E1
has a signed type and a negative value, the resulting value is
implementation-defined.
"""
Not only this, but (arithmetic) right shifting a negative value (using 2's
complement) is *not* equivalent to dividing it by the corresponding power
of 2. Let's look at a few examples:
-4 -> 0xF..FC
-4 >> 3 -> 0xF..FF == -1 != -4 / 8
-8 -> 0xF..F8
-8 >> 3 -> 0xF..FF == -1 == -8 / 8
-9 -> 0xF..F7
-9 >> 3 -> 0xF..FE == -2 != -9 / 8
Make update_avg() use a division, and export it to the private scheduler
header to reuse it where relevant. Note that this still lets compilers use
a shift here, but should prevent any unwanted surprise. The disassembly of
select_idle_cpu() remains unchanged on arm64, and ttwu_do_wakeup() gains 2
instructions; the diff sort of looks like this:
- sub x1, x1, x0
+ subs x1, x1, x0 // set condition codes
+ add x0, x1, #0x7
+ csel x0, x0, x1, mi // x0 = x1 < 0 ? x0 : x1
add x0, x3, x0, asr #3
which does the right thing (i.e. gives us the expected result while still
using an arithmetic shift)
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200330090127.16294-1-valentin.schneider@arm.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull core SMP updates from Thomas Gleixner:
"CPU (hotplug) updates:
- Support for locked CSD objects in smp_call_function_single_async()
which allows to simplify callsites in the scheduler core and MIPS
- Treewide consolidation of CPU hotplug functions which ensures the
consistency between the sysfs interface and kernel state. The low
level functions cpu_up/down() are now confined to the core code and
not longer accessible from random code"
* tag 'smp-core-2020-03-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (22 commits)
cpu/hotplug: Ignore pm_wakeup_pending() for disable_nonboot_cpus()
cpu/hotplug: Hide cpu_up/down()
cpu/hotplug: Move bringup of secondary CPUs out of smp_init()
torture: Replace cpu_up/down() with add/remove_cpu()
firmware: psci: Replace cpu_up/down() with add/remove_cpu()
xen/cpuhotplug: Replace cpu_up/down() with device_online/offline()
parisc: Replace cpu_up/down() with add/remove_cpu()
sparc: Replace cpu_up/down() with add/remove_cpu()
powerpc: Replace cpu_up/down() with add/remove_cpu()
x86/smp: Replace cpu_up/down() with add/remove_cpu()
arm64: hibernate: Use bringup_hibernate_cpu()
cpu/hotplug: Provide bringup_hibernate_cpu()
arm64: Use reboot_cpu instead of hardconding it to 0
arm64: Don't use disable_nonboot_cpus()
ARM: Use reboot_cpu instead of hardcoding it to 0
ARM: Don't use disable_nonboot_cpus()
ia64: Replace cpu_down() with smp_shutdown_nonboot_cpus()
cpu/hotplug: Create a new function to shutdown nonboot cpus
cpu/hotplug: Add new {add,remove}_cpu() functions
sched/core: Remove rq.hrtick_csd_pending
...
|
|
For simplicity, cpu pressure is defined as having more than one
runnable task on a given CPU. This works on the system-level, but it
has limitations in a cgrouped reality: When cpu.max is in use, it
doesn't capture the time in which a task is not executing on the CPU
due to throttling. Likewise, it doesn't capture the time in which a
competing cgroup is occupying the CPU - meaning it only reflects
cgroup-internal competitive pressure, not outside pressure.
Enable tracking of currently executing tasks, and then change the
definition of cpu pressure in a cgroup from
NR_RUNNING > 1
to
NR_RUNNING > ON_CPU
which will capture the effects of cpu.max as well as competition from
outside the cgroup.
After this patch, a cgroup running `stress -c 1` with a cpu.max
setting of 5000 10000 shows ~50% continuous CPU pressure.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200316191333.115523-2-hannes@cmpxchg.org
|
|
Currently, when updating the affinity of tasks via either cpusets.cpus,
or, sched_setaffinity(); tasks not currently running within the newly
specified mask will be arbitrarily assigned to the first CPU within the
mask.
This (particularly in the case that we are restricting masks) can
result in many tasks being assigned to the first CPUs of their new
masks.
This:
1) Can induce scheduling delays while the load-balancer has a chance to
spread them between their new CPUs.
2) Can antogonize a poor load-balancer behavior where it has a
difficult time recognizing that a cross-socket imbalance has been
forced by an affinity mask.
This change adds a new cpumask interface to allow iterated calls to
distribute within the intersection of the provided masks.
The cases that this mainly affects are:
- modifying cpuset.cpus
- when tasks join a cpuset
- when modifying a task's affinity via sched_setaffinity(2)
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Qais Yousef <qais.yousef@arm.com>
Tested-by: Qais Yousef <qais.yousef@arm.com>
Link: https://lkml.kernel.org/r/20200311010113.136465-1-joshdon@google.com
|
|
generic scheduler code
drivers/base/arch_topology.c is only built if CONFIG_GENERIC_ARCH_TOPOLOGY=y,
resulting in such build failures:
cpufreq_cooling.c:(.text+0x1e7): undefined reference to `arch_set_thermal_pressure'
Move it to sched/core.c instead, and keep it enabled on x86 despite
us not having a arch_scale_thermal_pressure() facility there, to
build-test this thing.
Cc: Thara Gopinath <thara.gopinath@linaro.org>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Now smp_call_function_single_async() provides the protection that
we'll return with -EBUSY if the csd object is still pending, then we
don't need the rq.hrtick_csd_pending any more.
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20191216213125.9536-4-peterx@redhat.com
|
|
Thermal pressure follows pelt signals which means the decay period for
thermal pressure is the default pelt decay period. Depending on SoC
characteristics and thermal activity, it might be beneficial to decay
thermal pressure slower, but still in-tune with the pelt signals. One way
to achieve this is to provide a command line parameter to set a decay
shift parameter to an integer between 0 and 10.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200222005213.3873-10-thara.gopinath@linaro.org
|
|
Introduce support in scheduler periodic tick and other CFS bookkeeping
APIs to trigger the process of computing average thermal pressure for a
CPU. Also consider avg_thermal.load_avg in others_have_blocked which
allows for decay of pelt signals.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20200222005213.3873-7-thara.gopinath@linaro.org
|
|
Now that runnable_load_avg is no more used, we can remove it to make
space for a new signal.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: "Dietmar Eggemann <dietmar.eggemann@arm.com>"
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Link: https://lore.kernel.org/r/20200224095223.13361-8-mgorman@techsingularity.net
|
|
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
A redundant "curr = rq->curr" was added; remove it.
Fixes: ebc0f83c78a2 ("timers/nohz: Update NOHZ load in remote tick")
Signed-off-by: Scott Wood <swood@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/1580776558-12882-1-git-send-email-swood@redhat.com
|
|
to fix XFS performance regression
The following XFS commit:
8ab39f11d974 ("xfs: prevent CIL push holdoff in log recovery")
changed the logic from using bound workqueues to using unbound
workqueues. Functionally this makes sense but it was observed at the
time that the dbench performance dropped quite a lot and CPU migrations
were increased.
The current pattern of the task migration is straight-forward. With XFS,
an IO issuer delegates work to xlog_cil_push_work ()on an unbound kworker.
This runs on a nearby CPU and on completion, dbench wakes up on its old CPU
as it is still idle and no migration occurs. dbench then queues the real
IO on the blk_mq_requeue_work() work item which runs on a bound kworker
which is forced to run on the same CPU as dbench. When IO completes,
the bound kworker wakes dbench but as the kworker is a bound but,
real task, the CPU is not considered idle and dbench gets migrated by
select_idle_sibling() to a new CPU. dbench may ping-pong between two CPUs
for a while but ultimately it starts a round-robin of all CPUs sharing
the same LLC. High-frequency migration on each IO completion has poor
performance overall. It has negative implications both in commication
costs and power management. mpstat confirmed that at low thread counts
that all CPUs sharing an LLC has low level of activity.
Note that even if the CIL patch was reverted, there still would
be migrations but the impact is less noticeable. It turns out that
individually the scheduler, XFS, blk-mq and workqueues all made sensible
decisions but in combination, the overall effect was sub-optimal.
This patch special cases the IO issue/completion pattern and allows
a bound kworker waker and a task wakee to stack on the same CPU if
there is a strong chance they are directly related. The expectation
is that the kworker is likely going back to sleep shortly. This is not
guaranteed as the IO could be queued asynchronously but there is a very
strong relationship between the task and kworker in this case that would
justify stacking on the same CPU instead of migrating. There should be
few concerns about kworker starvation given that the special casing is
only when the kworker is the waker.
DBench on XFS
MMTests config: io-dbench4-async modified to run on a fresh XFS filesystem
UMA machine with 8 cores sharing LLC
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack
Amean 1 22.63 ( 0.00%) 20.54 * 9.23%*
Amean 2 25.56 ( 0.00%) 23.40 * 8.44%*
Amean 4 28.63 ( 0.00%) 27.85 * 2.70%*
Amean 8 37.66 ( 0.00%) 37.68 ( -0.05%)
Amean 64 469.47 ( 0.00%) 468.26 ( 0.26%)
Stddev 1 1.00 ( 0.00%) 0.72 ( 28.12%)
Stddev 2 1.62 ( 0.00%) 1.97 ( -21.54%)
Stddev 4 2.53 ( 0.00%) 3.58 ( -41.19%)
Stddev 8 5.30 ( 0.00%) 5.20 ( 1.92%)
Stddev 64 86.36 ( 0.00%) 94.53 ( -9.46%)
NUMA machine, 48 CPUs total, 24 CPUs share cache
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack-v1r2
Amean 1 58.69 ( 0.00%) 30.21 * 48.53%*
Amean 2 60.90 ( 0.00%) 35.29 * 42.05%*
Amean 4 66.77 ( 0.00%) 46.55 * 30.28%*
Amean 8 81.41 ( 0.00%) 68.46 * 15.91%*
Amean 16 113.29 ( 0.00%) 107.79 * 4.85%*
Amean 32 199.10 ( 0.00%) 198.22 * 0.44%*
Amean 64 478.99 ( 0.00%) 477.06 * 0.40%*
Amean 128 1345.26 ( 0.00%) 1372.64 * -2.04%*
Stddev 1 2.64 ( 0.00%) 4.17 ( -58.08%)
Stddev 2 4.35 ( 0.00%) 5.38 ( -23.73%)
Stddev 4 6.77 ( 0.00%) 6.56 ( 3.00%)
Stddev 8 11.61 ( 0.00%) 10.91 ( 6.04%)
Stddev 16 18.63 ( 0.00%) 19.19 ( -3.01%)
Stddev 32 38.71 ( 0.00%) 38.30 ( 1.06%)
Stddev 64 100.28 ( 0.00%) 91.24 ( 9.02%)
Stddev 128 186.87 ( 0.00%) 160.34 ( 14.20%)
Dbench has been modified to report the time to complete a single "load
file". This is a more meaningful metric for dbench that a throughput
metric as the benchmark makes many different system calls that are not
throughput-related
Patch shows a 9.23% and 48.53% reduction in the time to process a load
file with the difference partially explained by the number of CPUs sharing
a LLC. In a separate run, task migrations were almost eliminated by the
patch for low client counts. In case people have issue with the metric
used for the benchmark, this is a comparison of the throughputs as
reported by dbench on the NUMA machine.
dbench4 Throughput (misleading but traditional)
5.5.0-rc7 5.5.0-rc7
tipsched-20200124 kworkerstack-v1r2
Hmean 1 321.41 ( 0.00%) 617.82 * 92.22%*
Hmean 2 622.87 ( 0.00%) 1066.80 * 71.27%*
Hmean 4 1134.56 ( 0.00%) 1623.74 * 43.12%*
Hmean 8 1869.96 ( 0.00%) 2212.67 * 18.33%*
Hmean 16 2673.11 ( 0.00%) 2806.13 * 4.98%*
Hmean 32 3032.74 ( 0.00%) 3039.54 ( 0.22%)
Hmean 64 2514.25 ( 0.00%) 2498.96 * -0.61%*
Hmean 128 1778.49 ( 0.00%) 1746.05 * -1.82%*
Note that this is somewhat specific to XFS and ext4 shows no performance
difference as it does not rely on kworkers in the same way. No major
problem was observed running other workloads on different machines although
not all tests have completed yet.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200128154006.GD3466@techsingularity.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Implement arch_scale_freq_capacity() for 'modern' x86. This function
is used by the scheduler to correctly account usage in the face of
DVFS.
The present patch addresses Intel processors specifically and has positive
performance and performance-per-watt implications for the schedutil cpufreq
governor, bringing it closer to, if not on-par with, the powersave governor
from the intel_pstate driver/framework.
Large performance gains are obtained when the machine is lightly loaded and
no regression are observed at saturation. The benchmarks with the largest
gains are kernel compilation, tbench (the networking version of dbench) and
shell-intensive workloads.
1. FREQUENCY INVARIANCE: MOTIVATION
* Without it, a task looks larger if the CPU runs slower
2. PECULIARITIES OF X86
* freq invariance accounting requires knowing the ratio freq_curr/freq_max
2.1 CURRENT FREQUENCY
* Use delta_APERF / delta_MPERF * freq_base (a.k.a "BusyMHz")
2.2 MAX FREQUENCY
* It varies with time (turbo). As an approximation, we set it to a
constant, i.e. 4-cores turbo frequency.
3. EFFECTS ON THE SCHEDUTIL FREQUENCY GOVERNOR
* The invariant schedutil's formula has no feedback loop and reacts faster
to utilization changes
4. KNOWN LIMITATIONS
* In some cases tasks can't reach max util despite how hard they try
5. PERFORMANCE TESTING
5.1 MACHINES
* Skylake, Broadwell, Haswell
5.2 SETUP
* baseline Linux v5.2 w/ non-invariant schedutil. Tested freq_max = 1-2-3-4-8-12
active cores turbo w/ invariant schedutil, and intel_pstate/powersave
5.3 BENCHMARK RESULTS
5.3.1 NEUTRAL BENCHMARKS
* NAS Parallel Benchmark (HPC), hackbench
5.3.2 NON-NEUTRAL BENCHMARKS
* tbench (10-30% better), kernbench (10-15% better),
shell-intensive-scripts (30-50% better)
* no regressions
5.3.3 SELECTION OF DETAILED RESULTS
5.3.4 POWER CONSUMPTION, PERFORMANCE-PER-WATT
* dbench (5% worse on one machine), kernbench (3% worse),
tbench (5-10% better), shell-intensive-scripts (10-40% better)
6. MICROARCH'ES ADDRESSED HERE
* Xeon Core before Scalable Performance processors line (Xeon Gold/Platinum
etc have different MSRs semantic for querying turbo levels)
7. REFERENCES
* MMTests performance testing framework, github.com/gormanm/mmtests
+-------------------------------------------------------------------------+
| 1. FREQUENCY INVARIANCE: MOTIVATION
+-------------------------------------------------------------------------+
For example; suppose a CPU has two frequencies: 500 and 1000 Mhz. When
running a task that would consume 1/3rd of a CPU at 1000 MHz, it would
appear to consume 2/3rd (or 66.6%) when running at 500 MHz, giving the
false impression this CPU is almost at capacity, even though it can go
faster [*]. In a nutshell, without frequency scale-invariance tasks look
larger just because the CPU is running slower.
[*] (footnote: this assumes a linear frequency/performance relation; which
everybody knows to be false, but given realities its the best approximation
we can make.)
+-------------------------------------------------------------------------+
| 2. PECULIARITIES OF X86
+-------------------------------------------------------------------------+
Accounting for frequency changes in PELT signals requires the computation of
the ratio freq_curr / freq_max. On x86 neither of those terms is readily
available.
2.1 CURRENT FREQUENCY
====================
Since modern x86 has hardware control over the actual frequency we run
at (because amongst other things, Turbo-Mode), we cannot simply use
the frequency as requested through cpufreq.
Instead we use the APERF/MPERF MSRs to compute the effective frequency
over the recent past. Also, because reading MSRs is expensive, don't
do so every time we need the value, but amortize the cost by doing it
every tick.
2.2 MAX FREQUENCY
=================
Obtaining freq_max is also non-trivial because at any time the hardware can
provide a frequency boost to a selected subset of cores if the package has
enough power to spare (eg: Turbo Boost). This means that the maximum frequency
available to a given core changes with time.
The approach taken in this change is to arbitrarily set freq_max to a constant
value at boot. The value chosen is the "4-cores (4C) turbo frequency" on most
microarchitectures, after evaluating the following candidates:
* 1-core (1C) turbo frequency (the fastest turbo state available)
* around base frequency (a.k.a. max P-state)
* something in between, such as 4C turbo
To interpret these options, consider that this is the denominator in
freq_curr/freq_max, and that ratio will be used to scale PELT signals such as
util_avg and load_avg. A large denominator will undershoot (util_avg looks a
bit smaller than it really is), viceversa with a smaller denominator PELT
signals will tend to overshoot. Given that PELT drives frequency selection
in the schedutil governor, we will have:
freq_max set to | effect on DVFS
--------------------+------------------
1C turbo | power efficiency (lower freq choices)
base freq | performance (higher util_avg, higher freq requests)
4C turbo | a bit of both
4C turbo proves to be a good compromise in a number of benchmarks (see below).
+-------------------------------------------------------------------------+
| 3. EFFECTS ON THE SCHEDUTIL FREQUENCY GOVERNOR
+-------------------------------------------------------------------------+
Once an architecture implements a frequency scale-invariant utilization (the
PELT signal util_avg), schedutil switches its frequency selection formula from
freq_next = 1.25 * freq_curr * util [non-invariant util signal]
to
freq_next = 1.25 * freq_max * util [invariant util signal]
where, in the second formula, freq_max is set to the 1C turbo frequency (max
turbo). The advantage of the second formula, whose usage we unlock with this
patch, is that freq_next doesn't depend on the current frequency in an
iterative fashion, but can jump to any frequency in a single update. This
absence of feedback in the formula makes it quicker to react to utilization
changes and more robust against pathological instabilities.
Compare it to the update formula of intel_pstate/powersave:
freq_next = 1.25 * freq_max * Busy%
where again freq_max is 1C turbo and Busy% is the percentage of time not spent
idling (calculated with delta_MPERF / delta_TSC); essentially the same as
invariant schedutil, and largely responsible for intel_pstate/powersave good
reputation. The non-invariant schedutil formula is derived from the invariant
one by approximating util_inv with util_raw * freq_curr / freq_max, but this
has limitations.
Testing shows improved performances due to better frequency selections when
the machine is lightly loaded, and essentially no change in behaviour at
saturation / overutilization.
+-------------------------------------------------------------------------+
| 4. KNOWN LIMITATIONS
+-------------------------------------------------------------------------+
It's been shown that it is possible to create pathological scenarios where a
CPU-bound task cannot reach max utilization, if the normalizing factor
freq_max is fixed to a constant value (see [Lelli-2018]).
If freq_max is set to 4C turbo as we do here, one needs to peg at least 5
cores in a package doing some busywork, and observe that none of those task
will ever reach max util (1024) because they're all running at less than the
4C turbo frequency.
While this concern still applies, we believe the performance benefit of
frequency scale-invariant PELT signals outweights the cost of this limitation.
[Lelli-2018]
https://lore.kernel.org/lkml/20180517150418.GF22493@localhost.localdomain/
+-------------------------------------------------------------------------+
| 5. PERFORMANCE TESTING
+-------------------------------------------------------------------------+
5.1 MACHINES
============
We tested the patch on three machines, with Skylake, Broadwell and Haswell
CPUs. The details are below, together with the available turbo ratios as
reported by the appropriate MSRs.
* 8x-SKYLAKE-UMA:
Single socket E3-1240 v5, Skylake 4 cores/8 threads
Max EFFiciency, BASE frequency and available turbo levels (MHz):
EFFIC 800 |********
BASE 3500 |***********************************
4C 3700 |*************************************
3C 3800 |**************************************
2C 3900 |***************************************
1C 3900 |***************************************
* 80x-BROADWELL-NUMA:
Two sockets E5-2698 v4, 2x Broadwell 20 cores/40 threads
Max EFFiciency, BASE frequency and available turbo levels (MHz):
EFFIC 1200 |************
BASE 2200 |**********************
8C 2900 |*****************************
7C 3000 |******************************
6C 3100 |*******************************
5C 3200 |********************************
4C 3300 |*********************************
3C 3400 |**********************************
2C 3600 |************************************
1C 3600 |************************************
* 48x-HASWELL-NUMA
Two sockets E5-2670 v3, 2x Haswell 12 cores/24 th |
