I noticed that alloc_bootmem_huge_page() will only advance to the next
node on failure to allocate a huge page, potentially filling nodes with
huge-pages.  I asked about this on linux-mm and linux-numa, cc'ing the
usual huge page suspects.

Mel Gorman responded:

	I strongly suspect that the same node being used until allocation
	failure instead of round-robin is an oversight and not deliberate
	at all. It appears to be a side-effect of a fix made way back in
	commit 63b4613c ["hugetlb: fix
	hugepage allocation with memoryless nodes"]. Prior to that patch
	it looked like allocations would always round-robin even when
	allocation was successful.

This patch--factored out of my "hugetlb mempolicy" series--moves the
advance of the hstate next node from which to allocate up before the test
for success of the attempted allocation.

Note that alloc_bootmem_huge_page() is only used for order > MAX_ORDER
huge pages.

I'll post a separate patch for mainline/stable, as the above mentioned
"balance freeing" series renamed the next node to alloc function.
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Reviewed-by: NMel Gorman <mel@csn.ul.ie>
Reviewed-by: NAndy Whitcroft <apw@canonical.com>
Reviewed-by: NAndi Kleen <andi@firstfloor.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Attempt to clarify huge page administration and usage, and updates the
doucmentation to mention the balancing of huge pages across nodes when
allocating and freeing.
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@canonical.com>
Cc: Eric Whitney <eric.whitney@hp.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use the [modified] free_pool_huge_page() function to return unused
surplus pages.  This will help keep huge pages balanced across nodes
between freeing of unused surplus pages and freeing of persistent huge
pages [from set_max_huge_pages] by using the same node id "cursor". It
also eliminates some code duplication.
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@canonical.com>
Cc: Eric Whitney <eric.whitney@hp.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Free huges pages from nodes in round robin fashion in an attempt to keep
[persistent a.k.a static] hugepages balanced across nodes

New function free_pool_huge_page() is modeled on and performs roughly the
inverse of alloc_fresh_huge_page().  Replaces dequeue_huge_page() which
now has no callers, so this patch removes it.

Helper function hstate_next_node_to_free() uses new hstate member
next_to_free_nid to distribute "frees" across all nodes with huge pages.
Acked-by: NDavid Rientjes <rientjes@google.com>
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: NMel Gorman <mel@csn.ul.ie>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@canonical.com>
Cc: Eric Whitney <eric.whitney@hp.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Ummark function as having kernel-doc notation, fixing the kernel-doc
warning.

Warning(mm/page_alloc.c:4519): No description found for parameter 'zone'
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In test, some pages in swap-cache can't be migrated, as they aren't rmap.

unmap_and_move() ignores swap-cache page which is just read in and hasn't
rmap (see the comments in the code), but swap_aops provides .migratepage.
Better to migrate such pages instead of ignore them.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Yakui Zhao <yakui.zhao@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

To initialize hotadded node, some pages are allocated.  At that time, the
node hasn't memory, this makes the allocation always fail.  In such case,
let's allocate pages from other nodes.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NYakui Zhao <yakui.zhao@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Pages on movable zone have two types, MIGRATE_MOVABLE and MIGRATE_RESERVE,
both them can be movable, because only movable memory allocation can get
pages from movable zone.  This makes pages in movable zone always be able
to migrate.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Yakui Zhao <yakui.zhao@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Pages marked as isolated should not be allocated again.  If such pages
reside in pcp list, they can be allocated too, so there is a ping-pong
memory offline frees some pages to pcp list and the pages get allocated
and then memory offline frees them again, this loop will happen again and
again.

This should have no impact in normal code path, because in normal code
path, pages in pcp list aren't isolated, and below loop will break in the
first entry.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Yakui Zhao <yakui.zhao@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In my test, 128M memory is hot added, but zone's pcp batch is 0, which is
an obvious error.  When pages are onlined, zone pcp should be updated
accordingly.

[akpm@linux-foundation.org: fix warnings]
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Yakui Zhao <yakui.zhao@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When a cpuset's nodemask is updated, all attached tasks have their cached
task->mems_allowed updated by a heap instead of requiring an explicit call
to cpuset_update_task_memory_state(), which has since been removed in
58568d2a ("cpuset,mm: update tasks'
mems_allowed in time").

Remove the obsoleted comment from the page allocator.

Cc: Paul Menage <menage@google.com>
Acked-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make use of the compiler's typechecking on !CONFIG_SWAP as well.

[akpm@linux-foundation.org: build fix]
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NRik van Riel <riel@redhat.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

drivers/mfd/ab3100-core.c:647: error: ab3100_init_settings causes a section type conflict

Cc: Anton Vorontsov <avorontsov@ru.mvista.com>
Cc: Samuel Ortiz <sameo@linux.intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We added a new column in cpuX lines of /proc/stat, to show the amount of
time spent by a cpu servicing a guest, without updating
Documentation/filesystems/proc.txt
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In theory it could happen that on one CPU we initialize a new inode but
clearing of I_NEW | I_LOCK gets reordered before some of the
initialization.  Thus on another CPU we return not fully uptodate inode
from iget_locked().

This seems to fix a corruption issue on ext3 mounted over NFS.

[akpm@linux-foundation.org: add some commentary]
Signed-off-by: NJan Kara <jack@suse.cz>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Remove myself as maintainer from the sdhci driver and steer people
towards the new MMC list for discussing it.
Signed-off-by: NPierre Ossman <pierre@ossman.eu>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

alpha:

drivers/media/dvb/pt1/pt1.c: In function 'pt1_cleanup_tables':
drivers/media/dvb/pt1/pt1.c:422: error: implicit declaration of function 'vfree'
drivers/media/dvb/pt1/pt1.c: In function 'pt1_init_tables':
drivers/media/dvb/pt1/pt1.c:431: error: implicit declaration of function 'vmalloc'
drivers/media/dvb/pt1/pt1.c:431: warning: assignment makes pointer from integer without a cast
Acked-by: NMauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Merge branch 'perfcounters-rename-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'perfcounters-rename-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  perf: Tidy up after the big rename
  perf: Do the big rename: Performance Counters -> Performance Events
  perf_counter: Rename 'event' to event_id/hw_event
  perf_counter: Rename list_entry -> group_entry, counter_list -> group_list

Manually resolved some fairly trivial conflicts with the tracing tree in
include/trace/ftrace.h and kernel/trace/trace_syscalls.c.

Merge branch 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  rcu: Fix whitespace inconsistencies
  rcu: Fix thinko, actually initialize full tree
  rcu: Apply results of code inspection of kernel/rcutree_plugin.h
  rcu: Add WARN_ON_ONCE() consistency checks covering state transitions
  rcu: Fix synchronize_rcu() for TREE_PREEMPT_RCU
  rcu: Simplify rcu_read_unlock_special() quiescent-state accounting
  rcu: Add debug checks to TREE_PREEMPT_RCU for premature grace periods
  rcu: Kconfig help needs to say that TREE_PREEMPT_RCU scales down
  rcutorture: Occasionally delay readers enough to make RCU force_quiescent_state
  rcu: Initialize multi-level RCU grace periods holding locks
  rcu: Need to update rnp->gpnum if preemptable RCU is to be reliable

Merge branch 'perfcounters-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'perfcounters-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  perf_counter, powerpc, sparc: Fix compilation after perf_counter_overflow() change
  perf_counter: x86: Fix PMU resource leak
  perf util: SVG performance improvements
  perf util: Make the timechart SVG width dynamic
  perf timechart: Show the duration of scheduler delays in the SVG
  perf timechart: Show the name of the waker/wakee in timechart

Merge branch 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  sched: Simplify sys_sched_rr_get_interval() system call
  sched: Fix potential NULL derference of doms_cur
  sched: Fix raciness in runqueue_is_locked()
  sched: Re-add lost cpu_allowed check to sched_fair.c::select_task_rq_fair()
  sched: Remove unneeded indentation in sched_fair.c::place_entity()

Merge branch 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  kernel/profile.c: Switch /proc/irq/prof_cpu_mask to seq_file
  tracing: Export trace_profile_buf symbols
  tracing/events: use list_for_entry_continue
  tracing: remove max_tracer_type_len
  function-graph: use ftrace_graph_funcs directly
  tracing: Remove markers
  tracing: Allocate the ftrace event profile buffer dynamically
  tracing: Factorize the events profile accounting

Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: Print the hypervisor returned tsc_khz during boot
  x86: Correct segment permission flags in 64-bit linker script
  x86: cpuinit-annotate SMP boot trampolines properly
  x86: Increase timeout for EHCI debug port reset completion in early printk
  x86: Fix uaccess_32.h typo
  x86: Trivial whitespace cleanups
  x86, apic: Fix missed handling of discrete apics
  x86/i386: Remove duplicated #include
  x86, mtrr: Convert loop to a while based construct, avoid naked semicolon
  Revert 'x86: Fix system crash when loading with "reservetop" parameter'
  x86, mce: Fix compile warning in case of CONFIG_SMP=n
  x86, apic: Use logical flat on intel with <= 8 logical cpus
  x86: SGI UV: Map MMIO-High memory range
  x86: SGI UV: Add volatile semantics to macros that access chipset registers
  x86: SGI UV: Fix IPI macros
  x86: apic: Convert BUG() to BUG_ON()
  x86: Remove final bits of CONFIG_X86_OLD_MCE

* 'writeback' of git://git.kernel.dk/linux-2.6-block:
  nfs: initialize the backing_dev_info when creating the server
  writeback: make balance_dirty_pages() gradually back more off
  writeback: don't use schedule_timeout() without setting runstate
  nfs: nfs_kill_super() should call bdi_unregister() after killing super

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6: (222 commits)
  V4L/DVB (13033): pt1: Don't use a deprecated DMA_BIT_MASK macro
  V4L/DVB (13029): radio-si4713: remove #include <linux/version.h>
  V4L/DVB (13027): go7007: convert printks to v4l2_info
  V4L/DVB (13026): s2250-board: Implement brightness and contrast controls
  V4L/DVB (13025): s2250-board: Fix memory leaks
  V4L/DVB (13024): go7007: Implement vidioc_g_std and vidioc_querystd
  V4L/DVB (13023): go7007: Merge struct gofh and go declarations
  V4L/DVB (13022): go7007: Fix mpeg controls
  V4L/DVB (13021): go7007: Fix whitespace and line lengths
  V4L/DVB (13020): go7007: Updates to Kconfig and Makefile
  V4L/DVB (13019): video: initial support for ADV7180
  V4L/DVB (13018): kzalloc failure ignored in au8522_probe()
  V4L/DVB (13017): gspca: kmalloc failure ignored in sd_start()
  V4L/DVB (13016): kmalloc failure ignored in lgdt3304_attach() and s921_attach()
  V4L/DVB (13015): kmalloc failure ignored in m920x_firmware_download()
  V4L/DVB (13014): Add support for Compro VideoMate E800 (DVB-T part only)
  V4L/DVB (13013): FM TX: si4713: Kconfig: Fixed two typos.
  V4L/DVB (13012): uvc: introduce missing kfree
  V4L/DVB (13011): Change tuner type of BeholdTV cards
  V4L/DVB (13009): gspca - stv06xx-hdcs: Reduce exposure range
  ...

* 'linux-next' of git://git.infradead.org/ubifs-2.6:
  UBIFS: fix debugging dump
  UBIFS: improve lprops dump
  UBIFS: various minor commentary fixes
  UBIFS: improve journal head debugging prints
  UBIFS: define journal head numbers in ubifs-media.h
  UBIFS: amend commentaries
  UBIFS: check ubifs_scan error codes better
  UBIFS: do not print scary error messages needlessly
  UBIFS: add inode size debugging check
  UBIFS: constify file and inode operations
  UBIFS: remove unneeded call from ubifs_sync_fs
  UBIFS: kill BKL
  UBIFS: remove unused functions
  UBIFS: suppress compilation warning

* 'linux-next' of git://git.infradead.org/ubi-2.6:
  UBI: improve NOR flash erasure quirk
  UBI: introduce flash dump helper
  UBI: eliminate possible undefined behaviour
  UBI: print a warning if too many PEBs are corrupted
  UBI: amend NOR flash pre-erase quirk
  UBI: print a message if ECH is corrupted and VIDH is ok

* 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: (133 commits)
  drm/vgaarb: add VGA arbitration support to the drm and kms.
  drm/radeon: some r420s have a CP race with the DMA engine.
  drm/radeon/r600/kms: rv670 is not DCE3
  drm/radeon/kms: r420 idle after programming GA_ENHANCE
  drm/radeon/kms: more fixes to rv770 suspend/resume path.
  drm/radeon/kms: more alignment for rv770.c with r600.c
  drm/radeon/kms: rv770 blit init called too late.
  drm/radeon/kms: move around new init path code to avoid posting at init
  drm/radeon/r600: fix some issues with suspend/resume.
  drm/radeon/kms: disable VGA rendering engine before taking over VRAM
  drm/radeon/kms: Move radeon_get_clock_info() call out of radeon_clocks_init().
  drm/radeon/kms: add initial connector properties
  drm/radeon/kms: Use surfaces for scanout / cursor byte swapping on big endian.
  drm/radeon/kms: don't fail if we fail to init GPU acceleration
  drm/r600/kms: fixup number of loops per blit calculation.
  drm/radeon/kms: reprogram format in set base.
  drm/radeon: avivo chips have no separate int bit for display
  drm/radeon/r600: don't do interrupts
  drm: fix _DRM_GEM addmap error message
  drm: update crtc x/y when only fb changes
  ...

Fixed up trivial conflicts in firmware/Makefile due to network driver
(cxgb3) and drm (mga/r128/radeon) firmware being listed next to each
other.

The build of the dabusb driver broke:

  drivers/media/video/dabusb.c:758: error: unknown field 'nodename' specified in initializer
  drivers/media/video/dabusb.c:758: warning: initialization from incompatible pointer type
  make[3]: *** wait: No child processes.  Stop.

Due to this commit:

  e454cea2: Driver-Core: extend devnode callbacks to provide permissions

Missing the dabusb driver's dabusb_nodename() callback.

Similar issues with the iio/industrialio driver in staging, pointed out
and patched by Jean Delvare.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Industrialio-parts-by: NJean Delvare <khali@linux-fr.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

NFS may free the server structure without ever having used the
bdi, so we either need to flag the bdi as being uninitialized or
initialize it up front. This does the latter.

This fixes a crash with mounting more than one NFS file system,
should people ever need that kind of obscure NFS functionality.
Tested-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Currently it just sleeps for a very short time, just 1 jiffy. If
we keep looping in there, continually delay for a little longer
of up to 100msec in total. That was the old limit for congestion
wait.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>