linux/fs/proc
Rik van Riel 34e431b0ae /proc/meminfo: provide estimated available memory
Many load balancing and workload placing programs check /proc/meminfo to
estimate how much free memory is available.  They generally do this by
adding up "free" and "cached", which was fine ten years ago, but is
pretty much guaranteed to be wrong today.

It is wrong because Cached includes memory that is not freeable as page
cache, for example shared memory segments, tmpfs, and ramfs, and it does
not include reclaimable slab memory, which can take up a large fraction
of system memory on mostly idle systems with lots of files.

Currently, the amount of memory that is available for a new workload,
without pushing the system into swap, can be estimated from MemFree,
Active(file), Inactive(file), and SReclaimable, as well as the "low"
watermarks from /proc/zoneinfo.

However, this may change in the future, and user space really should not
be expected to know kernel internals to come up with an estimate for the
amount of free memory.

It is more convenient to provide such an estimate in /proc/meminfo.  If
things change in the future, we only have to change it in one place.

Signed-off-by: Rik van Riel <riel@redhat.com>
Reported-by: Erik Mouw <erik.mouw_2@nxp.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-01-21 16:19:43 -08:00
..
array.c sched/numa: Report a NUMA task group ID 2013-10-09 14:47:49 +02:00
base.c audit: allow unsetting the loginuid (with priv) 2013-11-05 11:08:09 -05:00
cmdline.c
consoles.c seq_file: remove "%n" usage from seq_file users 2013-11-15 09:32:20 +09:00
cpuinfo.c
devices.c proc: use seq_puts()/seq_putc() where possible 2011-01-13 08:03:16 -08:00
fd.c proc: make proc_fd_permission() thread-friendly 2013-09-11 15:59:03 -07:00
fd.h proc: Move proc_fd() to fs/proc/fd.h 2013-05-01 17:29:39 -04:00
generic.c consolidate simple ->d_delete() instances 2013-11-15 22:04:17 -05:00
inode.c procfs: also fix proc_reg_get_unmapped_area() for !MMU case 2013-12-12 18:19:26 -08:00
internal.h proc_fill_cache(): just make instantiate_t return int 2013-06-29 12:57:18 +04:00
interrupts.c
Kconfig kcore: add Kconfig help text 2013-11-13 12:09:33 +09:00
kcore.c mm: use pgdat_end_pfn() to simplify the code in others 2013-11-13 12:09:03 +09:00
kmsg.c kmsg: honor dmesg_restrict sysctl on /dev/kmsg 2013-06-12 16:29:44 -07:00
loadavg.c sched, timers: cleanup avenrun users 2009-05-15 15:32:45 +02:00
Makefile mm, vmalloc: move get_vmalloc_info() to vmalloc.c 2013-04-29 15:54:33 -07:00
meminfo.c /proc/meminfo: provide estimated available memory 2014-01-21 16:19:43 -08:00
namespaces.c consolidate simple ->d_delete() instances 2013-11-15 22:04:17 -05:00
nommu.c seq_file: remove "%n" usage from seq_file users 2013-11-15 09:32:20 +09:00
page.c kpageflags: fix wrong KPF_THP on non-huge compound pages 2012-10-09 16:23:00 +09:00
proc_devtree.c of: remove HAVE_ARCH_DEVTREE_FIXUPS 2013-10-09 20:04:08 -05:00
proc_net.c [readdir] convert procfs 2013-06-29 12:56:32 +04:00
proc_sysctl.c Don't pass inode to ->d_hash() and ->d_compare() 2013-06-29 12:57:36 +04:00
proc_tty.c proc: use seq_puts()/seq_putc() where possible 2011-01-13 08:03:16 -08:00
root.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace 2013-09-07 14:35:32 -07:00
self.c new helper: kfree_put_link() 2013-10-24 23:34:49 -04:00
softirqs.c proc: use seq_puts()/seq_putc() where possible 2011-01-13 08:03:16 -08:00
stat.c stat: Use size_t for sizes instead of unsigned 2013-02-01 12:32:08 +02:00
task_mmu.c seq_file: remove "%n" usage from seq_file users 2013-11-15 09:32:20 +09:00
task_nommu.c seq_file: remove "%n" usage from seq_file users 2013-11-15 09:32:20 +09:00
uptime.c fs/proc/uptime.c:uptime_proc_show(): use get_monotonic_boottime() 2013-07-03 16:08:02 -07:00
version.c
vmcore.c vmcore: enable /proc/vmcore mmap for s390 2013-09-11 15:59:14 -07:00