kernel_optimize_test/kernel
Linus Torvalds 3510ca20ec Minor page waitqueue cleanups
Tim Chen and Kan Liang have been battling a customer load that shows
extremely long page wakeup lists.  The cause seems to be constant NUMA
migration of a hot page that is shared across a lot of threads, but the
actual root cause for the exact behavior has not been found.

Tim has a patch that batches the wait list traversal at wakeup time, so
that we at least don't get long uninterruptible cases where we traverse
and wake up thousands of processes and get nasty latency spikes.  That
is likely 4.14 material, but we're still discussing the page waitqueue
specific parts of it.

In the meantime, I've tried to look at making the page wait queues less
expensive, and failing miserably.  If you have thousands of threads
waiting for the same page, it will be painful.  We'll need to try to
figure out the NUMA balancing issue some day, in addition to avoiding
the excessive spinlock hold times.

That said, having tried to rewrite the page wait queues, I can at least
fix up some of the braindamage in the current situation. In particular:

 (a) we don't want to continue walking the page wait list if the bit
     we're waiting for already got set again (which seems to be one of
     the patterns of the bad load).  That makes no progress and just
     causes pointless cache pollution chasing the pointers.

 (b) we don't want to put the non-locking waiters always on the front of
     the queue, and the locking waiters always on the back.  Not only is
     that unfair, it means that we wake up thousands of reading threads
     that will just end up being blocked by the writer later anyway.

Also add a comment about the layout of 'struct wait_page_key' - there is
an external user of it in the cachefiles code that means that it has to
match the layout of 'struct wait_bit_key' in the two first members.  It
so happens to match, because 'struct page *' and 'unsigned long *' end
up having the same values simply because the page flags are the first
member in struct page.

Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Christopher Lameter <cl@linux.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-08-27 13:55:12 -07:00
..
bpf
cgroup cpuset: fix a deadlock due to incomplete patching of cpusets_enabled() 2017-08-02 17:16:12 -07:00
configs
debug
events perf/core: Fix group {cpu,task} validation 2017-08-25 11:00:34 +02:00
gcov
irq genirq/ipi: Fixup checks against nr_cpu_ids 2017-08-20 10:49:05 +02:00
livepatch
locking
power mm: fix global NR_SLAB_.*CLAIMABLE counter reads 2017-08-10 15:54:06 -07:00
printk
rcu
sched Minor page waitqueue cleanups 2017-08-27 13:55:12 -07:00
time timers: Fix excessive granularity of new timers after a nohz idle 2017-08-24 11:40:18 +02:00
trace Various bug fixes: 2017-08-24 14:08:22 -07:00
.gitignore
acct.c
async.c
audit_fsnotify.c
audit_tree.c
audit_watch.c audit/stable-4.13 PR 20170816 2017-08-16 16:48:34 -07:00
audit.c
audit.h
auditfilter.c
auditsc.c
backtracetest.c
bounds.c
capability.c
compat.c
configs.c
context_tracking.c
cpu_pm.c
cpu.c
crash_core.c
crash_dump.c
cred.c
delayacct.c
dma.c
elfcore.c
exec_domain.c
exit.c
extable.c
fork.c fork: fix incorrect fput of ->exe_file causing use-after-free 2017-08-25 16:12:46 -07:00
freezer.c
futex_compat.c
futex.c futex: Remove unnecessary warning from get_futex_key 2017-08-09 14:00:54 -07:00
groups.c
hung_task.c
irq_work.c
jump_label.c
kallsyms.c
kcmp.c
Kconfig.freezer
Kconfig.hz
Kconfig.locks
Kconfig.preempt
kcov.c
kexec_core.c
kexec_file.c
kexec_internal.h
kexec.c
kmod.c kmod: fix wait on recursive loop 2017-08-18 15:32:01 -07:00
kprobes.c
ksysfs.c
kthread.c
latencytop.c
Makefile
membarrier.c
memremap.c
module_signing.c
module-internal.h
module.c
notifier.c
nsproxy.c
padata.c
panic.c
params.c
pid_namespace.c
pid.c pids: make task_tgid_nr_ns() safe 2017-08-21 12:47:31 -07:00
profile.c
ptrace.c
range.c
reboot.c
relay.c
resource.c
seccomp.c
signal.c signal: don't remove SIGNAL_UNKILLABLE for traced tasks. 2017-08-18 15:32:02 -07:00
smp.c
smpboot.c
smpboot.h
softirq.c
stacktrace.c
stop_machine.c
sys_ni.c
sys.c
sysctl_binary.c
sysctl.c
task_work.c
taskstats.c
test_kprobes.c
torture.c
tracepoint.c
tsacct.c
ucount.c
uid16.c
up.c
user_namespace.c
user-return-notifier.c
user.c
utsname_sysctl.c
utsname.c
watchdog_hld.c kernel/watchdog: Prevent false positives with turbo modes 2017-08-18 12:35:02 +02:00
watchdog.c kernel/watchdog: Prevent false positives with turbo modes 2017-08-18 12:35:02 +02:00
workqueue_internal.h
workqueue.c