Merge remote-tracking branch 'tip/auto-latest'

Conflicts:
arch/Kconfig
arch/s390/Kconfig
arch/x86/Kconfig

Merge remote-tracking branch 'spi-mb/spi-next'

Merge remote-tracking branch 'dt-rh/for-next'

Merge remote-tracking branch 'edac-amd/for-next'

Conflicts:
Documentation/edac.txt
drivers/edac/amd64_edac.c

Merge remote-tracking branch 'edac/linux_next'

Merge remote-tracking branch 'fsnotify/for-next'

Conflicts:
kernel/audit_tree.c

Merge remote-tracking branch 'pm/linux-next'

Merge remote-tracking branch 'trivial/for-next'

Conflicts:
drivers/scsi/ipr.c
drivers/video/backlight/88pm860x_bl.c

Merge remote-tracking branch 'osd/linux-next'

Merge remote-tracking branch 'iommu/next'

Merge remote-tracking branch 'watchdog/master'

Merge remote-tracking branch 'selinux/master'

Merge remote-tracking branch 'security/next'

Merge remote-tracking branch 'regulator/for-next'

Conflicts:
Documentation/devicetree/bindings/regulator/tps6586x.txt

Merge remote-tracking branch 'omap_dss2/for-next'

Merge remote-tracking branch 'fbdev/fbdev-next'

Merge remote-tracking branch 'battery/master'

Conflicts:
include/linux/mfd/88pm860x.h

Merge remote-tracking branch 'mfd/for-next'

Conflicts:
Documentation/devicetree/bindings/regulator/tps6586x.txt
drivers/mfd/88pm860x-core.c
drivers/mfd/max8925-core.c
drivers/mfd/tc3589x.c
drivers/mfd/tps65217.c
drivers/regulator/anatop-regulator.c

Merge remote-tracking branch 'md/for-next'

Conflicts:
drivers/md/raid0.c
fs/bio.c

Merge remote-tracking branch 'slab/for-next'

Merge remote-tracking branch 'kgdb/kgdb-next'

Merge remote-tracking branch 'mmc/mmc-next'

Merge branch 'quilt/device-mapper'

Merge remote-tracking branch 'block/for-next'

Merge remote-tracking branch 'cgroup/for-next'

Merge remote-tracking branch 'input/next'

Merge branch 'quilt/rr'

Conflicts:
arch/xtensa/Kconfig

Merge remote-tracking branch 'sound-asoc/for-next'

Merge remote-tracking branch 'sound/for-next'

Conflicts:
arch/arm/mach-imx/mach-imx27_visstrim_m10.c

Merge remote-tracking branch 'drm/drm-next'

Conflicts:
drivers/gpu/drm/i915/i915_dma.c
drivers/gpu/drm/i915/intel_sdvo.c

Merge remote-tracking branch 'crypto/master'

Conflicts:
crypto/Kconfig

Merge remote-tracking branch 'l2-mtd/master'

Merge remote-tracking branch 'mtd/master'

Merge remote-tracking branch 'bluetooth/master'

Merge remote-tracking branch 'wireless-next/master'

Conflicts:
drivers/net/wireless/ath/ath9k/pci.c

Revert "be2net: fix vfs enumeration"

This reverts commit 51af6d7c1f31e0f3d42c87d53657ec7acb6e3462.

Merge remote-tracking branch 'net-next/master'

Conflicts:
drivers/net/team/team.c
drivers/net/usb/qmi_wwan.c
net/batman-adv/bat_iv_ogm.c
net/ipv4/fib_frontend.c
net/ipv4/route.c
net/l2tp/l2tp_netlink.c
net/socket.c

Merge remote-tracking branch 'slave-dma/next'

Merge remote-tracking branch 'target-updates/for-next'

Merge remote-tracking branch 'scsi/for-next'

Conflicts:
drivers/scsi/ipr.c
drivers/scsi/qla2xxx/qla_nx.c

Merge remote-tracking branch 'dlm/next'

Merge remote-tracking branch 'ubi/linux-next'

Merge remote-tracking branch 'ieee1394/for-next'

Merge remote-tracking branch 'thermal/next'

Merge remote-tracking branch 'cpuidle/cpuidle-next'

Conflicts:
drivers/cpuidle/coupled.c
include/linux/cpuidle.h

Merge remote-tracking branch 'acpi/next'

Merge remote-tracking branch 'pstore/master'

Merge remote-tracking branch 'infiniband/for-next'

Conflicts:
drivers/net/ethernet/mellanox/mlx4/mlx4.h

Merge remote-tracking branch 'libata/NEXT'

Merge remote-tracking branch 'kconfig/for-next'

Merge remote-tracking branch 'kbuild/for-next'

Merge remote-tracking branch 'v4l-dvb/master'

Conflicts:
arch/arm/mach-imx/mach-imx27_visstrim_m10.c

Merge remote-tracking branch 'hwmon-staging/hwmon-next'

Merge branch 'quilt/jdelvare-hwmon'

Merge remote-tracking branch 'i2c-embedded/i2c-embedded/for-next'

Merge branch 'quilt/i2c'

Merge remote-tracking branch 'hid/for-next'

Merge remote-tracking branch 'galak/next'

Merge remote-tracking branch 'powerpc/next'

Merge remote-tracking branch 'pci/next'

Merge remote-tracking branch 'vfs/for-next'

Merge remote-tracking branch 'xfs/for-next'

Merge remote-tracking branch 'ubifs/linux-next'

Merge remote-tracking branch 'v9fs/for-next'

Merge remote-tracking branch 'ocfs2/linux-next'

Merge remote-tracking branch 'nfsd/nfsd-next'

Merge remote-tracking branch 'nfs/linux-next'

Merge remote-tracking branch 'logfs/master'

Merge remote-tracking branch 'jfs/jfs-next'

Merge remote-tracking branch 'gfs2/master'

Merge remote-tracking branch 'fuse/for-next'

Merge remote-tracking branch 'ext4/dev'

Merge remote-tracking branch 'ext3/for_next'

Merge remote-tracking branch 'cifs/for-next'

Merge remote-tracking branch 'xtensa/for_next'

Merge remote-tracking branch 'unicore32/unicore32'

Merge remote-tracking branch 'sparc-next/master'

Merge remote-tracking branch 's390/features'

Merge remote-tracking branch 'openrisc/for-upstream'

Merge remote-tracking branch 'mips/mips-for-linux-next'

Merge remote-tracking branch 'm68knommu/for-next'

Merge remote-tracking branch 'm68k/for-next'

Merge remote-tracking branch 'ia64/next'

Merge remote-tracking branch 'cris/for-next'

Merge remote-tracking branch 'arm64/upstream'

Merge remote-tracking branch 'arm-perf/for-next/perf'

Conflicts:
arch/arm/kernel/hw_breakpoint.c

Merge remote-tracking branch 'arm/for-next'

Merge remote-tracking branch 'spi-current/spi/merge'

Merge remote-tracking branch 'devicetree-current/devicetree/merge'

Merge remote-tracking branch 'crypto-current/master'

Merge remote-tracking branch 'usb.current/usb-linus'

Merge remote-tracking branch 'net/master'

Merge remote-tracking branch 'sparc/master'

Merge remote-tracking branch 'm68k-current/for-linus'

Merge remote-tracking branch 'kbuild-current/rc-fixes'

virtio-blk: Disable callback in virtblk_done()

This reduces unnecessary interrupts that host could send to guest while
guest is in the progress of irq handling.

If one vcpu is handling the irq, while another interrupt comes, in
handle_edge_irq(), the guest will mask the interrupt via mask_msi_irq()
which is a very heavy operation that goes all the way down to host.

 Here are some performance numbers on qemu:

 Before:
 -------------------------------------
   seq-read  : io=0 B, bw=269730KB/s, iops=67432 , runt= 62200msec
   seq-write : io=0 B, bw=339716KB/s, iops=84929 , runt= 49386msec
   rand-read : io=0 B, bw=270435KB/s, iops=67608 , runt= 62038msec
   rand-write: io=0 B, bw=354436KB/s, iops=88608 , runt= 47335msec
     clat (usec): min=101 , max=138052 , avg=14822.09, stdev=11771.01
     clat (usec): min=96 , max=81543 , avg=11798.94, stdev=7735.60
     clat (usec): min=128 , max=140043 , avg=14835.85, stdev=11765.33
     clat (usec): min=109 , max=147207 , avg=11337.09, stdev=5990.35
   cpu          : usr=15.93%, sys=60.37%, ctx=7764972, majf=0, minf=54
   cpu          : usr=32.73%, sys=120.49%, ctx=7372945, majf=0, minf=1
   cpu          : usr=18.84%, sys=58.18%, ctx=7775420, majf=0, minf=1
   cpu          : usr=24.20%, sys=59.85%, ctx=8307886, majf=0, minf=0
   vdb: ios=8389107/8368136, merge=0/0, ticks=19457874/14616506,
 in_queue=34206098, util=99.68%
  43: interrupt in total: 887320
 fio --exec_prerun="echo 3 > /proc/sys/vm/drop_caches" --group_reporting
 --ioscheduler=noop --thread --bs=4k --size=512MB --direct=1 --numjobs=16
 --ioengine=libaio --iodepth=64 --loops=3 --ramp_time=0
 --filename=/dev/vdb --name=seq-read --stonewall --rw=read
 --name=seq-write --stonewall --rw=write --name=rnd-read --stonewall
 --rw=randread --name=rnd-write --stonewall --rw=randwrite

 After:
 -------------------------------------
   seq-read  : io=0 B, bw=309503KB/s, iops=77375 , runt= 54207msec
   seq-write : io=0 B, bw=448205KB/s, iops=112051 , runt= 37432msec
   rand-read : io=0 B, bw=311254KB/s, iops=77813 , runt= 53902msec
   rand-write: io=0 B, bw=377152KB/s, iops=94287 , runt= 44484msec
     clat (usec): min=81 , max=90588 , avg=12946.06, stdev=9085.94
     clat (usec): min=57 , max=72264 , avg=8967.97, stdev=5951.04
     clat (usec): min=29 , max=101046 , avg=12889.95, stdev=9067.91
     clat (usec): min=52 , max=106152 , avg=10660.56, stdev=4778.19
   cpu          : usr=15.05%, sys=57.92%, ctx=7710941, majf=0, minf=54
   cpu          : usr=26.78%, sys=101.40%, ctx=7387891, majf=0, minf=2
   cpu          : usr=19.03%, sys=58.17%, ctx=7681976, majf=0, minf=8
   cpu          : usr=24.65%, sys=58.34%, ctx=8442632, majf=0, minf=4
   vdb: ios=8389086/8361888, merge=0/0, ticks=17243780/12742010,
 in_queue=30078377, util=99.59%
  43: interrupt in total: 1259639
 fio --exec_prerun="echo 3 > /proc/sys/vm/drop_caches" --group_reporting
 --ioscheduler=noop --thread --bs=4k --size=512MB --direct=1 --numjobs=16
 --ioengine=libaio --iodepth=64 --loops=3 --ramp_time=0
 --filename=/dev/vdb --name=seq-read --stonewall --rw=read
 --name=seq-write --stonewall --rw=write --name=rnd-read --stonewall
 --rw=randread --name=rnd-write --stonewall --rw=randwrite

Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>

Use the recently-added bio front_pad field to allocate struct dm_target_io.

Prior to this patch, dm_target_io was allocated from a mempool. For each
dm_target_io, there is exactly one bio allocated from a bioset.

This patch merges these two allocations into one allocation: we create a
bioset with front_pad equal to the size of dm_target_io so that every
bio allocated from the bioset has sizeof(struct dm_target_io) bytes
before it. We allocate a bio and use the bytes before the bio as
dm_target_io.

This idea was introduced by Kent Overstreet.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: Kent Overstreet <koverstreet@google.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: tj@kernel.org
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>

Support discards when the pool's block size is not a power of 2.
The block layer assumes discard_granularity is a power of 2 (in
blkdev_issue_discard), so we set this to the largest power of 2 that is
a divides into the number of sectors in each block, but never less than
DATA_DEV_BLOCK_SIZE_MIN_SECTORS.

This patch eliminates the "Discard support must be disabled when the
block size is not a power of 2" constraint that was imposed in commit
55f2b8b ("dm thin: support for non power of 2 pool blocksize").  That
commit was incomplete: using a block size that is not a power of 2
shouldn't mean disabling discard support on the device completely.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>

Use the ACCESS_ONCE macro in dm-bufio and dm-verity where a variable
can be modified asynchronously (through sysfs) and we want to prevent
compiler optimizations that assume that the variable hasn't changed.
(See Documentation/atomic_ops.txt.)

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>

Use list_move() instead of list_del() + list_add().

spatch with a semantic match was used to find this.
(http://coccinelle.lip6.fr/)

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>