Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 31 7月, 2020 2 次提交
  3. 29 7月, 2020 1 次提交
    • Q
      sched/uclamp: Add a new sysctl to control RT default boost value · 13685c4a
      由 Qais Yousef 提交于 
      RT tasks by default run at the highest capacity/performance level. When
uclamp is selected this default behavior is retained by enforcing the
requested uclamp.min (p->uclamp_req[UCLAMP_MIN]) of the RT tasks to be
uclamp_none(UCLAMP_MAX), which is SCHED_CAPACITY_SCALE; the maximum
value.

This is also referred to as 'the default boost value of RT tasks'.

See commit 1a00d999 ("sched/uclamp: Set default clamps for RT tasks").

On battery powered devices, it is desired to control this default
(currently hardcoded) behavior at runtime to reduce energy consumed by
RT tasks.

For example, a mobile device manufacturer where big.LITTLE architecture
is dominant, the performance of the little cores varies across SoCs, and
on high end ones the big cores could be too power hungry.

Given the diversity of SoCs, the new knob allows manufactures to tune
the best performance/power for RT tasks for the particular hardware they
run on.

They could opt to further tune the value when the user selects
a different power saving mode or when the device is actively charging.

The runtime aspect of it further helps in creating a single kernel image
that can be run on multiple devices that require different tuning.

Keep in mind that a lot of RT tasks in the system are created by the
kernel. On Android for instance I can see over 50 RT tasks, only
a handful of which created by the Android framework.

To control the default behavior globally by system admins and device
integrator, introduce the new sysctl_sched_uclamp_util_min_rt_default
to change the default boost value of the RT tasks.

I anticipate this to be mostly in the form of modifying the init script
of a particular device.

To avoid polluting the fast path with unnecessary code, the approach
taken is to synchronously do the update by traversing all the existing
tasks in the system. This could race with a concurrent fork(), which is
dealt with by introducing sched_post_fork() function which will ensure
the racy fork will get the right update applied.

Tested on Juno-r2 in combination with the RT capacity awareness [1].
By default an RT task will go to the highest capacity CPU and run at the
maximum frequency, which is particularly energy inefficient on high end
mobile devices because the biggest core[s] are 'huge' and power hungry.

With this patch the RT task can be controlled to run anywhere by
default, and doesn't cause the frequency to be maximum all the time.
Yet any task that really needs to be boosted can easily escape this
default behavior by modifying its requested uclamp.min value
(p->uclamp_req[UCLAMP_MIN]) via sched_setattr() syscall.

[1] 804d402f: ("sched/rt: Make RT capacity-aware")
Signed-off-by: NQais Yousef <qais.yousef@arm.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200716110347.19553-2-qais.yousef@arm.com
      13685c4a
  5. 28 7月, 2020 1 次提交
  7. 22 7月, 2020 1 次提交
  9. 10 7月, 2020 1 次提交
  11. 08 7月, 2020 3 次提交
  13. 04 7月, 2020 1 次提交
  15. 28 6月, 2020 2 次提交
  17. 11 6月, 2020 1 次提交
    • T
      x86/{mce,mm}: Unmap the entire page if the whole page is affected and poisoned · 17fae129
      由 Tony Luck 提交于 
      An interesting thing happened when a guest Linux instance took a machine
check. The VMM unmapped the bad page from guest physical space and
passed the machine check to the guest.

Linux took all the normal actions to offline the page from the process
that was using it. But then guest Linux crashed because it said there
was a second machine check inside the kernel with this stack trace:

do_memory_failure
    set_mce_nospec
         set_memory_uc
              _set_memory_uc
                   change_page_attr_set_clr
                        cpa_flush
                             clflush_cache_range_opt

This was odd, because a CLFLUSH instruction shouldn't raise a machine
check (it isn't consuming the data). Further investigation showed that
the VMM had passed in another machine check because is appeared that the
guest was accessing the bad page.

Fix is to check the scope of the poison by checking the MCi_MISC register.
If the entire page is affected, then unmap the page. If only part of the
page is affected, then mark the page as uncacheable.

This assumes that VMMs will do the logical thing and pass in the "whole
page scope" via the MCi_MISC register (since they unmapped the entire
page).

  [ bp: Adjust to x86/entry changes. ]

Fixes: 284ce401 ("x86/memory_failure: Introduce {set, clear}_mce_nospec()")
Reported-by: NJue Wang <juew@google.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>
Signed-off-by: NBorislav Petkov <bp@suse.de>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Tested-by: NJue Wang <juew@google.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20200520163546.GA7977@agluck-desk2.amr.corp.intel.com
      17fae129
  19. 05 6月, 2020 1 次提交
  21. 03 6月, 2020 1 次提交
    • N
      mm/writeback: replace PF_LESS_THROTTLE with PF_LOCAL_THROTTLE · a37b0715
      由 NeilBrown 提交于 
      PF_LESS_THROTTLE exists for loop-back nfsd (and a similar need in the
loop block driver and callers of prctl(PR_SET_IO_FLUSHER)), where a
daemon needs to write to one bdi (the final bdi) in order to free up
writes queued to another bdi (the client bdi).

The daemon sets PF_LESS_THROTTLE and gets a larger allowance of dirty
pages, so that it can still dirty pages after other processses have been
throttled.  The purpose of this is to avoid deadlock that happen when
the PF_LESS_THROTTLE process must write for any dirty pages to be freed,
but it is being thottled and cannot write.

This approach was designed when all threads were blocked equally,
independently on which device they were writing to, or how fast it was.
Since that time the writeback algorithm has changed substantially with
different threads getting different allowances based on non-trivial
heuristics.  This means the simple "add 25%" heuristic is no longer
reliable.

The important issue is not that the daemon needs a *larger* dirty page
allowance, but that it needs a *private* dirty page allowance, so that
dirty pages for the "client" bdi that it is helping to clear (the bdi
for an NFS filesystem or loop block device etc) do not affect the
throttling of the daemon writing to the "final" bdi.

This patch changes the heuristic so that the task is not throttled when
the bdi it is writing to has a dirty page count below below (or equal
to) the free-run threshold for that bdi.  This ensures it will always be
able to have some pages in flight, and so will not deadlock.

In a steady-state, it is expected that PF_LOCAL_THROTTLE tasks might
still be throttled by global threshold, but that is acceptable as it is
only the deadlock state that is interesting for this flag.

This approach of "only throttle when target bdi is busy" is consistent
with the other use of PF_LESS_THROTTLE in current_may_throttle(), were
it causes attention to be focussed only on the target bdi.

So this patch
 - renames PF_LESS_THROTTLE to PF_LOCAL_THROTTLE,
 - removes the 25% bonus that that flag gives, and
 - If PF_LOCAL_THROTTLE is set, don't delay at all unless the
   global and the local free-run thresholds are exceeded.

Note that previously realtime threads were treated the same as
PF_LESS_THROTTLE threads.  This patch does *not* change the behvaiour
for real-time threads, so it is now different from the behaviour of nfsd
and loop tasks.  I don't know what is wanted for realtime.

[akpm@linux-foundation.org: coding style fixes]
Signed-off-by: NNeilBrown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Acked-by: Chuck Lever <chuck.lever@oracle.com>	[nfsd]
Cc: Christoph Hellwig <hch@lst.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Link: http://lkml.kernel.org/r/87ftbf7gs3.fsf@notabene.neil.brown.nameSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a37b0715
  23. 28 5月, 2020 1 次提交
    • P
      sched: Replace rq::wake_list · a1488664
      由 Peter Zijlstra 提交于 
      The recent commit: 90b5363a ("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 ttwu_queue_remote() got this wrong.

While on first reading ttwu_queue_remote() 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.

The actual race is vs the sched_ttwu_pending() call in the idle loop;
that can run the wakeup-list without consuming the CSD.

Instead of trying to chain the lists, merge them.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20200526161908.129371594@infradead.org
      a1488664
  25. 19 5月, 2020 2 次提交
  27. 12 5月, 2020 1 次提交
  29. 28 4月, 2020 3 次提交
    • P
      rcu-tasks: Split ->trc_reader_need_end · 276c4104
      由 Paul E. McKenney 提交于 
      This commit splits ->trc_reader_need_end by using the rcu_special union.
This change permits readers to check to see if a memory barrier is
required without any added overhead in the common case where no such
barrier is required.  This commit also adds the read-side checking.
Later commits will add the machinery to properly set the new
->trc_reader_special.b.need_mb field.

This commit also makes rcu_read_unlock_trace_special() tolerate nested
read-side critical sections within interrupt and NMI handlers.
Signed-off-by: NPaul E. McKenney <paulmck@kernel.org>
      276c4104
    • P
      rcu-tasks: Add an RCU Tasks Trace to simplify protection of tracing hooks · d5f177d3
      由 Paul E. McKenney 提交于 
      Because RCU does not watch exception early-entry/late-exit, idle-loop,
or CPU-hotplug execution, protection of tracing and BPF operations is
needlessly complicated.  This commit therefore adds a variant of
Tasks RCU that:

o	Has explicit read-side markers to allow finite grace periods in
	the face of in-kernel loops for PREEMPT=n builds.  These markers
	are rcu_read_lock_trace() and rcu_read_unlock_trace().

o	Protects code in the idle loop, exception entry/exit, and
	CPU-hotplug code paths.  In this respect, RCU-tasks trace is
	similar to SRCU, but with lighter-weight readers.

o	Avoids expensive read-side instruction, having overhead similar
	to that of Preemptible RCU.

There are of course downsides:

o	The grace-period code can send IPIs to CPUs, even when those
	CPUs are in the idle loop or in nohz_full userspace.  This is
	mitigated by later commits.

o	It is necessary to scan the full tasklist, much as for Tasks RCU.

o	There is a single callback queue guarded by a single lock,
	again, much as for Tasks RCU.  However, those early use cases
	that request multiple grace periods in quick succession are
	expected to do so from a single task, which makes the single
	lock almost irrelevant.  If needed, multiple callback queues
	can be provided using any number of schemes.

Perhaps most important, this variant of RCU does not affect the vanilla
flavors, rcu_preempt and rcu_sched.  The fact that RCU Tasks Trace
readers can operate from idle, offline, and exception entry/exit in no
way enables rcu_preempt and rcu_sched readers to do so.

The memory ordering was outlined here:
https://lore.kernel.org/lkml/20200319034030.GX3199@paulmck-ThinkPad-P72/

This effort benefited greatly from off-list discussions of BPF
requirements with Alexei Starovoitov and Andrii Nakryiko.  At least
some of the on-list discussions are captured in the Link: tags below.
In addition, KCSAN was quite helpful in finding some early bugs.

Link: https://lore.kernel.org/lkml/20200219150744.428764577@infradead.org/
Link: https://lore.kernel.org/lkml/87mu8p797b.fsf@nanos.tec.linutronix.de/
Link: https://lore.kernel.org/lkml/20200225221305.605144982@linutronix.de/
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andrii Nakryiko <andriin@fb.com>
[ paulmck: Apply feedback from Steve Rostedt and Joel Fernandes. ]
[ paulmck: Decrement trc_n_readers_need_end upon IPI failure. ]
[ paulmck: Fix locking issue reported by rcutorture. ]
Signed-off-by: NPaul E. McKenney <paulmck@kernel.org>
      d5f177d3
    • L
      rcu: Remove unused ->rcu_read_unlock_special.b.deferred_qs field · f0bdf6d4
      由 Lai Jiangshan 提交于 
      The ->rcu_read_unlock_special.b.deferred_qs field is set to true in
rcu_read_unlock_special() but never set to false.  This is not
particularly useful, so this commit removes this field.

The only possible justification for this field is to ease debugging
of RCU deferred quiscent states, but the combination of the other
->rcu_read_unlock_special fields plus ->rcu_blocked_node and of course
->rcu_read_lock_nesting should cover debugging needs.  And if this last
proves incorrect, this patch can always be reverted, along with the
required setting of ->rcu_read_unlock_special.b.deferred_qs to false
in rcu_preempt_deferred_qs_irqrestore().
Signed-off-by: NLai Jiangshan <laijs@linux.alibaba.com>
Signed-off-by: NPaul E. McKenney <paulmck@kernel.org>
      f0bdf6d4
  31. 02 4月, 2020 1 次提交
    • E
      signal: Extend exec_id to 64bits · d1e7fd64
      由 Eric W. Biederman 提交于 
      Replace the 32bit exec_id with a 64bit exec_id to make it impossible
to wrap the exec_id counter.  With care an attacker can cause exec_id
wrap and send arbitrary signals to a newly exec'd parent.  This
bypasses the signal sending checks if the parent changes their
credentials during exec.

The severity of this problem can been seen that in my limited testing
of a 32bit exec_id it can take as little as 19s to exec 65536 times.
Which means that it can take as little as 14 days to wrap a 32bit
exec_id.  Adam Zabrocki has succeeded wrapping the self_exe_id in 7
days.  Even my slower timing is in the uptime of a typical server.
Which means self_exec_id is simply a speed bump today, and if exec
gets noticably faster self_exec_id won't even be a speed bump.

Extending self_exec_id to 64bits introduces a problem on 32bit
architectures where reading self_exec_id is no longer atomic and can
take two read instructions.  Which means that is is possible to hit
a window where the read value of exec_id does not match the written
value.  So with very lucky timing after this change this still
remains expoiltable.

I have updated the update of exec_id on exec to use WRITE_ONCE
and the read of exec_id in do_notify_parent to use READ_ONCE
to make it clear that there is no locking between these two
locations.

Link: https://lore.kernel.org/kernel-hardening/20200324215049.GA3710@pi3.com.pl
Fixes: 2.3.23pre2
Cc: stable@vger.kernel.org
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
      d1e7fd64
  33. 21 3月, 2020 2 次提交
    • S
      lockdep: Add hrtimer context tracing bits · 40db1739
      由 Sebastian Andrzej Siewior 提交于 
      Set current->irq_config = 1 for hrtimers which are not marked to expire in
hard interrupt context during hrtimer_init(). These timers will expire in
softirq context on PREEMPT_RT.

Setting this allows lockdep to differentiate these timers. If a timer is
marked to expire in hard interrupt context then the timer callback is not
supposed to acquire a regular spinlock instead of a raw_spinlock in the
expiry callback.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200321113242.534508206@linutronix.de
      40db1739
    • P
      lockdep: Introduce wait-type checks · de8f5e4f
      由 Peter Zijlstra 提交于 
      Extend lockdep to validate lock wait-type context.

The current wait-types are:

	LD_WAIT_FREE,		/* wait free, rcu etc.. */
	LD_WAIT_SPIN,		/* spin loops, raw_spinlock_t etc.. */
	LD_WAIT_CONFIG,		/* CONFIG_PREEMPT_LOCK, spinlock_t etc.. */
	LD_WAIT_SLEEP,		/* sleeping locks, mutex_t etc.. */

Where lockdep validates that the current lock (the one being acquired)
fits in the current wait-context (as generated by the held stack).

This ensures that there is no attempt to acquire mutexes while holding
spinlocks, to acquire spinlocks while holding raw_spinlocks and so on. In
other words, its a more fancy might_sleep().

Obviously RCU made the entire ordeal more complex than a simple single
value test because RCU can be acquired in (pretty much) any context and
while it presents a context to nested locks it is not the same as it
got acquired in.

Therefore its necessary to split the wait_type into two values, one
representing the acquire (outer) and one representing the nested context
(inner). For most 'normal' locks these two are the same.

[ To make static initialization easier we have the rule that:
  .outer == INV means .outer == .inner; because INV == 0. ]

It further means that its required to find the minimal .inner of the held
stack to compare against the outer of the new lock; because while 'normal'
RCU presents a CONFIG type to nested locks, if it is taken while already
holding a SPIN type it obviously doesn't relax the rules.

Below is an example output generated by the trivial test code:

  raw_spin_lock(&foo);
  spin_lock(&bar);
  spin_unlock(&bar);
  raw_spin_unlock(&foo);

 [ BUG: Invalid wait context ]
 -----------------------------
 swapper/0/1 is trying to lock:
 ffffc90000013f20 (&bar){....}-{3:3}, at: kernel_init+0xdb/0x187
 other info that might help us debug this:
 1 lock held by swapper/0/1:
  #0: ffffc90000013ee0 (&foo){+.+.}-{2:2}, at: kernel_init+0xd1/0x187

The way to read it is to look at the new -{n,m} part in the lock
description; -{3:3} for the attempted lock, and try and match that up to
the held locks, which in this case is the one: -{2,2}.

This tells that the acquiring lock requires a more relaxed environment than
presented by the lock stack.

Currently only the normal locks and RCU are converted, the rest of the
lockdep users defaults to .inner = INV which is ignored. More conversions
can be done when desired.

The check for spinlock_t nesting is not enabled by default. It's a separate
config option for now as there are known problems which are currently
addressed. The config option allows to identify these problems and to
verify that the solutions found are indeed solving them.

The config switch will be removed and the checks will permanently enabled
once the vast majority of issues has been addressed.

[ bigeasy: Move LD_WAIT_FREE,… out of CONFIG_LOCKDEP to avoid compile
	   failure with CONFIG_DEBUG_SPINLOCK + !CONFIG_LOCKDEP]
[ tglx: Add the config option ]
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200321113242.427089655@linutronix.de
      de8f5e4f
  35. 20 3月, 2020 1 次提交
  37. 24 2月, 2020 2 次提交
  39. 26 1月, 2020 1 次提交
  41. 25 12月, 2019 1 次提交
    • M
      rseq: Unregister rseq for clone CLONE_VM · 463f550f
      由 Mathieu Desnoyers 提交于 
      It has been reported by Google that rseq is not behaving properly
with respect to clone when CLONE_VM is used without CLONE_THREAD.

It keeps the prior thread's rseq TLS registered when the TLS of the
thread has moved, so the kernel can corrupt the TLS of the parent.

The approach of clearing the per task-struct rseq registration
on clone with CLONE_THREAD flag is incomplete. It does not cover
the use-case of clone with CLONE_VM set, but without CLONE_THREAD.

Here is the rationale for unregistering rseq on clone with CLONE_VM
flag set:

1) CLONE_THREAD requires CLONE_SIGHAND, which requires CLONE_VM to be
   set. Therefore, just checking for CLONE_VM covers all CLONE_THREAD
   uses. There is no point in checking for both CLONE_THREAD and
   CLONE_VM,

2) There is the possibility of an unlikely scenario where CLONE_SETTLS
   is used without CLONE_VM. In order to be an issue, it would require
   that the rseq TLS is in a shared memory area.

   I do not plan on adding CLONE_SETTLS to the set of clone flags which
   unregister RSEQ, because it would require that we also unregister RSEQ
   on set_thread_area(2) and arch_prctl(2) ARCH_SET_FS for completeness.
   So rather than doing a partial solution, it appears better to let
   user-space explicitly perform rseq unregistration across clone if
   needed in scenarios where CLONE_VM is not set.
Signed-off-by: NMathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191211161713.4490-3-mathieu.desnoyers@efficios.comSigned-off-by: NIngo Molnar <mingo@kernel.org>
      463f550f
  43. 05 12月, 2019 1 次提交
    • A
      kcov: remote coverage support · eec028c9
      由 Andrey Konovalov 提交于 
      Patch series " kcov: collect coverage from usb and vhost", v3.

This patchset extends kcov to allow collecting coverage from backgound
kernel threads.  This extension requires custom annotations for each of
the places where coverage collection is desired.  This patchset
implements this for hub events in the USB subsystem and for vhost
workers.  See the first patch description for details about the kcov
extension.  The other two patches apply this kcov extension to USB and
vhost.

Examples of other subsystems that might potentially benefit from this
when custom annotations are added (the list is based on
process_one_work() callers for bugs recently reported by syzbot):

1. fs: writeback wb_workfn() worker,
2. net: addrconf_dad_work()/addrconf_verify_work() workers,
3. net: neigh_periodic_work() worker,
4. net/p9: p9_write_work()/p9_read_work() workers,
5. block: blk_mq_run_work_fn() worker.

These patches have been used to enable coverage-guided USB fuzzing with
syzkaller for the last few years, see the details here:

  https://github.com/google/syzkaller/blob/master/docs/linux/external_fuzzing_usb.md

This patchset has been pushed to the public Linux kernel Gerrit
instance:

  https://linux-review.googlesource.com/c/linux/kernel/git/torvalds/linux/+/1524

This patch (of 3):

Add background thread coverage collection ability to kcov.

With KCOV_ENABLE coverage is collected only for syscalls that are issued
from the current process.  With KCOV_REMOTE_ENABLE it's possible to
collect coverage for arbitrary parts of the kernel code, provided that
those parts are annotated with kcov_remote_start()/kcov_remote_stop().

This allows to collect coverage from two types of kernel background
threads: the global ones, that are spawned during kernel boot in a
limited number of instances (e.g.  one USB hub_event() worker thread is
spawned per USB HCD); and the local ones, that are spawned when a user
interacts with some kernel interface (e.g.  vhost workers).

To enable collecting coverage from a global background thread, a unique
global handle must be assigned and passed to the corresponding
kcov_remote_start() call.  Then a userspace process can pass a list of
such handles to the KCOV_REMOTE_ENABLE ioctl in the handles array field
of the kcov_remote_arg struct.  This will attach the used kcov device to
the code sections, that are referenced by those handles.

Since there might be many local background threads spawned from
different userspace processes, we can't use a single global handle per
annotation.  Instead, the userspace process passes a non-zero handle
through the common_handle field of the kcov_remote_arg struct.  This
common handle gets saved to the kcov_handle field in the current
task_struct and needs to be passed to the newly spawned threads via
custom annotations.  Those threads should in turn be annotated with
kcov_remote_start()/kcov_remote_stop().

Internally kcov stores handles as u64 integers.  The top byte of a
handle is used to denote the id of a subsystem that this handle belongs
to, and the lower 4 bytes are used to denote the id of a thread instance
within that subsystem.  A reserved value 0 is used as a subsystem id for
common handles as they don't belong to a particular subsystem.  The
bytes 4-7 are currently reserved and must be zero.  In the future the
number of bytes used for the subsystem or handle ids might be increased.

When a particular userspace process collects coverage by via a common
handle, kcov will collect coverage for each code section that is
annotated to use the common handle obtained as kcov_handle from the
current task_struct.  However non common handles allow to collect
coverage selectively from different subsystems.

Link: http://lkml.kernel.org/r/e90e315426a384207edbec1d6aa89e43008e4caf.1572366574.git.andreyknvl@google.comSigned-off-by: NAndrey Konovalov <andreyknvl@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: David Windsor <dwindsor@gmail.com>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Anders Roxell <anders.roxell@linaro.org>
Cc: Alexander Potapenko <glider@google.com>
Cc: Marco Elver <elver@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      eec028c9
  45. 20 11月, 2019 2 次提交
  47. 16 11月, 2019 1 次提交
  49. 13 11月, 2019 1 次提交
  51. 30 10月, 2019 1 次提交
    • J
      io-wq: small threadpool implementation for io_uring · 771b53d0
      由 Jens Axboe 提交于 
      This adds support for io-wq, a smaller and specialized thread pool
implementation. This is meant to replace workqueues for io_uring. Among
the reasons for this addition are:

- We can assign memory context smarter and more persistently if we
  manage the life time of threads.

- We can drop various work-arounds we have in io_uring, like the
  async_list.

- We can implement hashed work insertion, to manage concurrency of
  buffered writes without needing a) an extra workqueue, or b)
  needlessly making the concurrency of said workqueue very low
  which hurts performance of multiple buffered file writers.

- We can implement cancel through signals, for cancelling
  interruptible work like read/write (or send/recv) to/from sockets.

- We need the above cancel for being able to assign and use file tables
  from a process.

- We can implement a more thorough cancel operation in general.

- We need it to move towards a syslet/threadlet model for even faster
  async execution. For that we need to take ownership of the used
  threads.

This list is just off the top of my head. Performance should be the
same, or better, at least that's what I've seen in my testing. io-wq
supports basic NUMA functionality, setting up a pool per node.

io-wq hooks up to the scheduler schedule in/out just like workqueue
and uses that to drive the need for more/less workers.
Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      771b53d0
  53. 29 10月, 2019 3 次提交
  55. 17 10月, 2019 1 次提交
    • J
      arm64: entry.S: Do not preempt from IRQ before all cpufeatures are enabled · 19c95f26
      由 Julien Thierry 提交于 
      Preempting from IRQ-return means that the task has its PSTATE saved
on the stack, which will get restored when the task is resumed and does
the actual IRQ return.

However, enabling some CPU features requires modifying the PSTATE. This
means that, if a task was scheduled out during an IRQ-return before all
CPU features are enabled, the task might restore a PSTATE that does not
include the feature enablement changes once scheduled back in.

* Task 1:

PAN == 0 ---|                          |---------------
            |                          |<- return from IRQ, PSTATE.PAN = 0
            | <- IRQ                   |
            +--------+ <- preempt()  +--
                                     ^
                                     |
                                     reschedule Task 1, PSTATE.PAN == 1
* Init:
        --------------------+------------------------
                            ^
                            |
                            enable_cpu_features
                            set PSTATE.PAN on all CPUs

Worse than this, since PSTATE is untouched when task switching is done,
a task missing the new bits in PSTATE might affect another task, if both
do direct calls to schedule() (outside of IRQ/exception contexts).

Fix this by preventing preemption on IRQ-return until features are
enabled on all CPUs.

This way the only PSTATE values that are saved on the stack are from
synchronous exceptions. These are expected to be fatal this early, the
exception is BRK for WARN_ON(), but as this uses do_debug_exception()
which keeps IRQs masked, it shouldn't call schedule().
Signed-off-by: NJulien Thierry <julien.thierry@arm.com>
[james: Replaced a really cool hack, with an even simpler static key in C.
 expanded commit message with Julien's cover-letter ascii art]
Signed-off-by: NJames Morse <james.morse@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>
      19c95f26