linux-stable/fs/proc
Vlastimil Babka 6a15a37097 mm, proc: reduce cost of /proc/pid/smaps for shmem mappings
The previous patch has improved swap accounting for shmem mapping, which
however made /proc/pid/smaps more expensive for shmem mappings, as we
consult the radix tree for each pte_none entry, so the overal complexity
is O(n*log(n)).

We can reduce this significantly for mappings that cannot contain COWed
pages, because then we can either use the statistics tha shmem object
itself tracks (if the mapping contains the whole object, or the swap
usage of the whole object is zero), or use the radix tree iterator,
which is much more effective than repeated find_get_entry() calls.

This patch therefore introduces a function shmem_swap_usage(vma) and
makes /proc/pid/smaps use it when possible.  Only for writable private
mappings of shmem objects (i.e.  tmpfs files) with the shmem object
itself (partially) swapped outwe have to resort to the find_get_entry()
approach.

Hopefully such mappings are relatively uncommon.

To demonstrate the diference, I have measured this on a process that
creates a 2GB mapping and dirties single pages with a stride of 2MB, and
time how long does it take to cat /proc/pid/smaps of this process 100
times.

Private writable mapping of a /dev/shm/file (the most complex case):

real    0m3.831s
user    0m0.180s
sys     0m3.212s

Shared mapping of an almost full mapping of a partially swapped /dev/shm/file
(which needs to employ the radix tree iterator).

real    0m1.351s
user    0m0.096s
sys     0m0.768s

Same, but with /dev/shm/file not swapped (so no radix tree walk needed)

real    0m0.935s
user    0m0.128s
sys     0m0.344s

Private anonymous mapping:

real    0m0.949s
user    0m0.116s
sys     0m0.348s

The cost is now much closer to the private anonymous mapping case, unless
the shmem mapping is private and writable.

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-01-14 16:00:49 -08:00
..
array.c fs/proc/array.c: set overflow flag in case of error 2015-11-06 17:50:42 -08:00
base.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-01-12 17:11:47 -08:00
cmdline.c
consoles.c
cpuinfo.c
devices.c
fd.c proc: add a reschedule point in proc_readfd_common() 2016-01-09 02:56:10 -05:00
fd.h
generic.c proc: change proc_subdir_lock to a rwlock 2015-09-10 13:29:01 -07:00
inode.c kmemcg: account certain kmem allocations to memcg 2016-01-14 16:00:49 -08:00
internal.h proc: Allow creating permanently empty directories that serve as mount points 2015-07-01 10:36:41 -05:00
interrupts.c
Kconfig fs, proc: add help for CONFIG_PROC_CHILDREN 2015-07-17 16:39:52 -07:00
kcore.c x86/fpu, sched: Introduce CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT and use it on x86 2015-07-18 03:42:51 +02:00
kmsg.c
loadavg.c
Makefile proc: Implement /proc/thread-self to point at the directory of the current thread 2014-08-04 10:07:11 -07:00
meminfo.c mm: get rid of 'vmalloc_info' from /proc/meminfo 2015-11-01 17:09:15 -08:00
namespaces.c switch ->get_link() to delayed_call, kill ->put_link() 2015-12-30 13:01:03 -05:00
nommu.c vfs: add seq_file_path() helper 2015-06-23 18:01:07 -04:00
page.c proc: add cond_resched to /proc/kpage* read/write loop 2015-09-10 13:29:01 -07:00
proc_net.c VFS: normal filesystems (and lustre): d_inode() annotations 2015-04-15 15:06:57 -04:00
proc_sysctl.c fs: Drop unlikely before IS_ERR(_OR_NULL) 2015-09-29 15:13:58 +02:00
proc_tty.c proc: remove proc_tty_ldisc variable 2014-08-08 15:57:22 -07:00
root.c vfs: Commit to never having exectuables on proc and sysfs. 2015-07-10 10:39:25 -05:00
self.c switch ->get_link() to delayed_call, kill ->put_link() 2015-12-30 13:01:03 -05:00
softirqs.c
stat.c genirq: Prevent proc race against freeing of irq descriptors 2014-12-13 13:33:07 +01:00
task_mmu.c mm, proc: reduce cost of /proc/pid/smaps for shmem mappings 2016-01-14 16:00:49 -08:00
task_nommu.c vfs: add seq_file_path() helper 2015-06-23 18:01:07 -04:00
thread_self.c switch ->get_link() to delayed_call, kill ->put_link() 2015-12-30 13:01:03 -05:00
uptime.c
version.c
vmcore.c vmcore: fix PT_NOTE n_namesz, n_descsz overflow issue 2015-02-17 14:34:52 -08:00