linux/fs/proc
KOSAKI Motohiro 00a62ce91e mm: fix Committed_AS underflow on large NR_CPUS environment
The Committed_AS field can underflow in certain situations:

>         # while true; do cat /proc/meminfo  | grep _AS; sleep 1; done | uniq -c
>               1 Committed_AS: 18446744073709323392 kB
>              11 Committed_AS: 18446744073709455488 kB
>               6 Committed_AS:    35136 kB
>               5 Committed_AS: 18446744073709454400 kB
>               7 Committed_AS:    35904 kB
>               3 Committed_AS: 18446744073709453248 kB
>               2 Committed_AS:    34752 kB
>               9 Committed_AS: 18446744073709453248 kB
>               8 Committed_AS:    34752 kB
>               3 Committed_AS: 18446744073709320960 kB
>               7 Committed_AS: 18446744073709454080 kB
>               3 Committed_AS: 18446744073709320960 kB
>               5 Committed_AS: 18446744073709454080 kB
>               6 Committed_AS: 18446744073709320960 kB

Because NR_CPUS can be greater than 1000 and meminfo_proc_show() does
not check for underflow.

But NR_CPUS proportional isn't good calculation.  In general,
possibility of lock contention is proportional to the number of online
cpus, not theorical maximum cpus (NR_CPUS).

The current kernel has generic percpu-counter stuff.  using it is right
way.  it makes code simplify and percpu_counter_read_positive() don't
make underflow issue.

Reported-by: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Eric B Munson <ebmunson@us.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: <stable@kernel.org>		[All kernel versions]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-02 15:36:10 -07:00
..
array.c
base.c proc: mounts_poll() make consistent to mdstat_poll 2009-04-16 16:17:10 -07:00
cmdline.c
cpuinfo.c
devices.c
generic.c proc 1/2: do PDE usecounting even for ->read_proc, ->write_proc 2009-03-31 01:14:27 +04:00
inode.c proc 2/2: remove struct proc_dir_entry::owner 2009-03-31 01:14:44 +04:00
internal.h proc 1/2: do PDE usecounting even for ->read_proc, ->write_proc 2009-03-31 01:14:27 +04:00
interrupts.c
Kconfig
kcore.c
kmsg.c
loadavg.c
Makefile
meminfo.c mm: fix Committed_AS underflow on large NR_CPUS environment 2009-05-02 15:36:10 -07:00
mmu.c
nommu.c trivial: fix typo "kernal" -> "kernel" 2009-03-30 15:21:57 +02:00
page.c proc: fix kflags to uflags copying in /proc/kpageflags 2009-03-11 07:43:33 -07:00
proc_devtree.c
proc_net.c
proc_sysctl.c constify dentry_operations: procfs 2009-03-27 14:44:01 -04:00
proc_tty.c proc tty: remove struct tty_operations::read_proc 2009-04-01 08:59:10 -07:00
root.c vfs: simple_set_mnt() should return void 2009-03-27 14:44:03 -04:00
stat.c [S390] /proc/stat idle field for idle cpus 2009-04-23 13:58:17 +02:00
task_mmu.c pagemap: require aligned-length, non-null reads of /proc/pid/pagemap 2009-05-02 15:36:09 -07:00
task_nommu.c nommu: fix typo vma->pg_off to vma->vm_pgoff 2009-04-08 10:21:44 -07:00
uptime.c Revert "proc: revert /proc/uptime to ->read_proc hook" 2009-03-31 01:14:58 +04:00
version.c
vmcore.c