2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-24 05:04:00 +08:00
linux-next/kernel/sched
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
..
autogroup.c sched/autogroup: Rename auto_group.[ch] to autogroup.[ch] 2017-02-08 09:01:11 +01:00
autogroup.h sched/headers: Prepare for new header dependencies before moving code to <linux/sched/autogroup.h> 2017-03-02 08:42:28 +01:00
clock.c sched/clock: Fix early boot preempt assumption in __set_sched_clock_stable() 2017-05-24 09:10:00 +02:00
completion.c sched/wait: Rename wait_queue_t => wait_queue_entry_t 2017-06-20 12:18:27 +02:00
core.c sched/core: Fix some documentation build warnings 2017-07-25 11:17:02 +02:00
cpuacct.c sched/cputime: Convert kcpustat to nsecs 2017-02-01 09:13:47 +01:00
cpuacct.h sched/cpuacct: Simplify the cpuacct code 2016-03-21 11:00:28 +01:00
cpudeadline.c sched/core: Remove the tsk_cpus_allowed() wrapper 2017-03-02 08:42:24 +01:00
cpudeadline.h sched/deadline: Split cpudl_set() into cpudl_set() and cpudl_clear() 2016-09-05 13:29:43 +02:00
cpufreq_schedutil.c cpufreq: schedutil: Fix sugov_start() versus sugov_update_shared() race 2017-07-12 14:47:48 +02:00
cpufreq.c cpufreq / sched: Pass flags to cpufreq_update_util() 2016-08-16 22:14:55 +02:00
cpupri.c sched/core: Remove the tsk_cpus_allowed() wrapper 2017-03-02 08:42:24 +01:00
cpupri.h sched/cpupri: Remove unnecessary definitions in cpupri.h 2014-11-16 10:58:59 +01:00
cputime.c sched/cputime: Don't use smp_processor_id() in preemptible context 2017-07-14 10:27:15 +02:00
deadline.c sched/deadline: Fix confusing comments about selection of top pi-waiter 2017-07-14 10:35:16 +02:00
debug.c sched/debug: Expose the number of RT/DL tasks that can migrate 2017-06-30 09:32:07 +02:00
fair.c sched/fair: Fix load_balance() affinity redo path 2017-07-05 16:28:48 +02:00
features.h sched/core: Implement new approach to scale select_idle_cpu() 2017-06-08 10:25:17 +02:00
idle_task.c sched/core: Add wrappers for lockdep_(un)pin_lock() 2017-01-14 11:29:30 +01:00
idle.c sched/idle: Add deferrable vmstat_updater back 2017-06-08 10:32:09 +02:00
loadavg.c sched/loadavg: Generalize "_idle" naming to "_nohz" 2017-06-22 11:30:01 +02:00
Makefile Merge branch 'WIP.sched/core' into sched/core 2017-06-20 12:28:21 +02:00
rt.c sched/rt: Move RT related code from sched/core.c to sched/rt.c 2017-06-23 10:46:45 +02:00
sched-pelt.h sched/fair: Move the PELT constants into a generated header 2017-04-14 10:26:37 +02:00
sched.h sched/rt: Move RT related code from sched/core.c to sched/rt.c 2017-06-23 10:46:45 +02:00
stats.c sched: use %*pb[l] to print bitmaps including cpumasks and nodemasks 2015-02-13 21:21:37 -08:00
stats.h sched/headers: Move cputime functionality from <linux/sched.h> and <linux/cputime.h> into <linux/sched/cputime.h> 2017-03-03 01:45:22 +01:00
stop_task.c sched/core: Add wrappers for lockdep_(un)pin_lock() 2017-01-14 11:29:30 +01:00
swait.c sched/headers: Prepare to move signal wakeup & sigpending methods from <linux/sched.h> into <linux/sched/signal.h> 2017-03-02 08:42:32 +01:00
topology.c sched/topology: Rename sched_group_cpus() 2017-05-15 10:15:34 +02:00
wait_bit.c sched/wait: Disambiguate wq_entry->task_list and wq_head->task_list naming 2017-06-20 12:19:14 +02:00
wait.c Minor page waitqueue cleanups 2017-08-27 13:55:12 -07:00