This is an improved straw bucket that correctly avoids any data movement
between items A and B when neither A nor B's weights are changed.  Said
differently, if we adjust the weight of item C (including adding it anew
or removing it completely), we will only see inputs move to or from C,
never between other items in the bucket.

Notably, there is not intermediate scaling factor that needs to be
calculated.  The mapping function is a simple function of the item weights.

The below commits were squashed together into this one (mostly to avoid
adding and then yanking a ~6000 lines worth of crush_ln_table):

- crush: add a straw2 bucket type
- crush: add crush_ln to calculate nature log efficently
- crush: improve straw2 adjustment slightly
- crush: change crush_ln to provide 32 more digits
- crush: fix crush_get_bucket_item_weight and bucket destroy for straw2
- crush/mapper: fix divide-by-0 in straw2
  (with div64_s64() for draw = ln / w and INT64_MIN -> S64_MIN - need
   to create a proper compat.h in ceph.git)

Reflects ceph.git commits 242293c908e923d474910f2b8203fa3b41eb5a53,
                          32a1ead92efcd351822d22a5fc37d159c65c1338,
                          6289912418c4a3597a11778bcf29ed5415117ad9,
                          35fcb04e2945717cf5cfe150b9fa89cb3d2303a1,
                          6445d9ee7290938de1e4ee9563912a6ab6d8ee5f,
                          b5921d55d16796e12d66ad2c4add7305f9ce2353.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Crush temporary buffers are allocated as per replica size configured
by the user.  When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash.  Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by the rule.

Reflects ceph.git commits 6b4d1aa99718e3b367496326c1e64551330fabc0,
                          234b066ba04976783d15ff2abc3e81b6cc06fb10.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

For CEPH_OSD_CMPXATTR_MODE_U64, OSD expects the u64 to be encoded
as string in object's xattr.
Signed-off-by: Yan, Zheng <zyan@redhat.com>

Signed-off-by: Yan, Zheng <zyan@redhat.com>

sb->s_root can be null when umounting
Signed-off-by: Yan, Zheng <zyan@redhat.com>

If a directory is complete, we want to keep the exclusive
cap. So that MDS does not end up revoking the shared cap
on every create/unlink operation.
Signed-off-by: Yan, Zheng <zyan@redhat.com>

No need to do single_open()'s job ourselves.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Don't pollute /proc/mounts with default options (presently these are
dcache, nofsc and acl).  Leave the acl/noacl however - it's a bit of
a special case due to CONFIG_CEPH_FS_POSIX_ACL.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Add a client_options attribute for showing libceph options.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Split ceph_show_options() into two pieces and move the piece
responsible for printing client (libceph) options into net/ceph.  This
way people adding a libceph option wouldn't have to remember to update
code in fs/ceph.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Set QUEUE_FLAG_NONROT.  Following commit b277da0a ("block: disable
entropy contributions for nonrot devices") we should also clear
QUEUE_FLAG_ADD_RANDOM, but it's off by default for blk-mq drivers, so
just note it in the comment.

Also remove physical block size assignment - no sense in repeating
defaults that are not going to change.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

- specific con->error_msg messages (e.g. "protocol version mismatch")
  end up getting overwritten by a catch-all "socket error on read
  / write", introduced in commit 3a140a0d ("libceph: report socket
  read/write error message")
- "bad message sequence # for incoming message" loses to "bad crc" due
  to the fact that -EBADMSG is used for both

Fix it, and tidy up con->error_msg assignments and pr_errs while at it.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Signed-off-by: Yan, Zheng <zyan@redhat.com>

remove_session_caps_cb() does not truncate dirty data in page
cache, but zeros i_wrbuffer_ref/i_wrbuffer_ref_head. This will
result negtive i_wrbuffer_ref/i_wrbuffer_ref_head
Signed-off-by: Yan, Zheng <zyan@redhat.com>

No i_auth_cap means reconnecting to MDS was denied. So don't
add new dirty caps.
Signed-off-by: Yan, Zheng <zyan@redhat.com>

when reconnecting to MDS is denied, we remove session caps
forcibly. But it's possible there are ongoing write, the
write code needs to reference i_snap_realm. So if there are
ongoing write, we keep i_snap_realm.
Signed-off-by: Yan, Zheng <zyan@redhat.com>

To avoid possible interleaving, add missing '\n' to formats.

Convert pr_warning to pr_warn while there.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Currently, there is no check for the kstrdup() for r_path2,
r_path1 and snapdir_name as various locations as there is a
possibility of failure during memory pressure. Therefore,
returning ENOMEM where the checks have been missed.
Signed-off-by: Sanidhya Kashyap <sanidhya.gatech@gmail.com>
Signed-off-by: Yan, Zheng <zyan@redhat.com>

When ceph_update_writeable_page fails (including -EAGAIN), it
unlocks (w/ unlock_page) the page but does not 'release'
(w/ page_cache_release) properly.

Upon error, properly set *pagep to NULL, indicating an error.
Signed-off-by: Taesoo Kim <tsgatesv@gmail.com>
Signed-off-by: Yan, Zheng <zyan@redhat.com>

pr_info what exactly was the culprit: missing pool, image or snap.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

return type of wait_for_completion_timeout is unsigned long not int. An
appropriately named unsigned long is added and the assignment fixed up.
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: Yan, Zheng <zyan@redhat.com>

This is only an API consolidation and should make things more readable
it replaces var * HZ / 1000 by msecs_to_jiffies(var).
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: Yan, Zheng <zyan@redhat.com>

ceph_aops was already defined extern in addr.c section
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Yan, Zheng <zyan@redhat.com>

Signed-off-by: Yan, Zheng <zyan@redhat.com>

These cap releases are stale because MDS will re-establish client
caps according to the cap reconnect messages.

Note: MDS can detect stale cap messages, so these stale cap
releases are harmless even we don't drop them.
Signed-off-by: Yan, Zheng <zyan@redhat.com>

Pull vfs and fs fixes from Al Viro:
 "Several AIO and OCFS2 fixes"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  ocfs2: _really_ sync the right range
  ocfs2_file_write_iter: keep return value and current position update in sync
  [regression] ocfs2: do *not* increment ->ki_pos twice
  ioctx_alloc(): fix vma (and file) leak on failure
  fix mremap() vs. ioctx_kill() race

Pull last minute thermal-SoC management fixes from Eduardo Valentin:
 "Specifics:

   - Minor fixes on ST and RCAR thermal drivers.
   - Avoid flooding kernel log when driver returns -EAGAIN.

  Note: I am sending this pull on Rui's behalf while he fixes issues in
  his Linux box"

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/evalenti/linux-soc-thermal:
  drivers: thermal: st: remove several sparse warnings
  thermal: constify of_device_id array
  thermal: Do not log an error if thermal_zone_get_temp returns -EAGAIN
  thermal: rcar: Fix typo in r8a73a4 SoC name

Pull last-minute ASoC fix from Mark Brown:
 "This patch backs out a change that came in during the merge window
  which selects a configuration for GPIO4 on pcm512x CODECs that may not
  be suitable for all systems using the device.  Changes for v4.1 will
  make this properly configurable but for now it's safest to revert to
  the v3.19 behaviour and leave the pin configuration alone.

  Sorry for sending this direct at the last minute but due to the GPIO
  misuse it'd be really good to get it in the release and I'd not
  realised it hadn't been sent yet - between some travel, a job change
  and other non-urgent fixes coming in I'd lost track of the urgency.

  It's been in -next for several weeks now, is isolated to the driver
  and fairly clear to inspection"

* tag 'asoc-fix-v4.0-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound:
  ASoC: pcm512x: Remove hardcoding of pll-lock to GPIO4

Currently GPIO4 is hardcoded to output the pll-lock signal.
Unfortunately this is after the pll-out GPIO is configured which
is selectable in the device tree. Therefore it is not possible to
use GPIO4 for pll-out. Therefore this patch removes the
configuration of GPIO4.
Signed-off-by: Howard Mitchell <hm@hmbedded.co.uk>
Signed-off-by: Mark Brown <broonie@kernel.org>

This reverts commit ecc19d17.

It added a new warning to try to encourage driver writers to set the
device capabities properly, but drivers haven't been updated and in the
meantime it just generaters a scary message that users cannot actually
do anything about.

Warnings like these are appropriate if you actually expect to fix the
code that causes them.  They are not appropriate for releases.
Requested-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Jan Engelhardt reports a strange oops with an invalid ->sense_buffer
pointer in scsi_init_cmd_errh() with the blk-mq code.

The sense_buffer pointer should have been initialized by the call to
scsi_init_request() from blk_mq_init_rq_map(), but there seems to be
some non-repeatable memory corruptor.

This patch makes sure we initialize the whole struct request allocation
(and the associated 'struct scsi_cmnd' for the SCSI case) to zero, by
using __GFP_ZERO in the allocation.  The old code initialized a couple
of individual fields, leaving the rest undefined (although many of them
are then initialized in later phases, like blk_mq_rq_ctx_init() etc.

It's not entirely clear why this matters, but it's the rigth thing to do
regardless, and with 4.0 imminent this is the defensive "let's just make
sure everything is initialized properly" patch.
Tested-by: Jan Engelhardt <jengelh@inai.de>
Acked-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull dmaengine fix from Vinod Koul:
 "I have one more fix to fix the boot warning on cppi driver due to
  missing capabilities"

* 'fixes' of git://git.infradead.org/users/vkoul/slave-dma:
  dmaengine: cppi41: add missing bitfields

Pull late ipmi fixes from Corey Minyard:
 "Some annoying issues in the IPMI driver that would be good to have
  fixed before 4.0 is released.

  These got reported or discovered late, but they will avoid some
  situations that would cause lots of log spam and in one case a
  deadlock"

* tag 'for-linus-4.0-1' of git://git.code.sf.net/p/openipmi/linux-ipmi:
  ipmi_ssif: Use interruptible completion for waiting in the thread
  ipmi/powernv: Fix minor locking bug
  ipmi: Handle BMCs that don't allow clearing the rcv irq bit

Add missing directions, residue_granularity,
srd_addr_widths and dst_addr_widths bitfields.

Without those we will see a kernel WARN()
when loading musb on am335x devices.
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>

The code was using an normal completion, but that caused stuck
task errors after a while.  Use an interruptible one to avoid that.
Signed-off-by: Corey Minyard <cminyard@mvista.com>

If ipmi_powernv_recv(...) is called without a current message it
prints a warning and returns. However it fails to release the message
lock causing the system to dead lock during any subsequent IPMI
operations.

This error path should never normally be taken unless there are bugs
elsewhere in the system.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Corey Minyard <cminyard@mvista.com>

Some BMCs don't let you clear the receive irq bit in the global
enables.  This is kind of silly, but they give an error if you
try to clear it.  Compensate for this by detecting the situation
and working around it.
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Tested-by: Thomas D <whissi@whissi.de>
Reviewed-by: Thomas D <whissi@whissi.de>

Pull SCSI fixes from James Bottomley:
 "This is our remaining set of three fixes for 4.0: two oops fixes(one
  for cable pulls triggering oopses and the other be2iscsi specific) and
  one warn on in sysfs on multipath devices using enclosures"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  Defer processing of REQ_PREEMPT requests for blocked devices
  be2iscsi: Fix kernel panic when device initialization fails
  enclosure: fix WARN_ON removing an adapter in multi-path devices