Although synchronize_sched_expedited() historically has no RCU CPU stall
warnings, the availability of the rcupdate.rcu_expedited boot parameter
invalidates the old assumption that synchronize_sched()'s stall warnings
would suffice.  This commit therefore adds RCU CPU stall warnings to
synchronize_sched_expedited().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The strictly rcu_node based funnel-locking scheme works well in many
cases, but systems with CONFIG_RCU_FANOUT_LEAF=64 won't necessarily get
all that much concurrency.  This commit therefore extends the funnel
locking into the per-CPU rcu_data structure, providing concurrency equal
to the number of CPUs.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

One of the requirements on RCU grace periods is that if there is a
causal chain of operations that starts after one grace period and
ends before another grace period, then the two grace periods must
be serialized.  There has been (and might still be) code that relies
on this, for example, certain types of reference-counting code that
does a call_rcu() within an RCU callback function.

This requirement is why there is an smp_mb() at the end of both
synchronize_sched_expedited() and synchronize_rcu_expedited().
However, this is the only smp_mb() in these functions, so it would
be nicer to consolidate it into rcu_exp_gp_seq_end().  This commit
does just that.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The rcu_seq operations were open-coded in _rcu_barrier(), so this commit
replaces the open-coding with the shiny new rcu_seq operations.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit gets rid of synchronize_rcu_expedited()'s mutex_trylock()
polling loop in favor of the funnel-locking scheme that was abstracted
from synchronize_sched_expedited().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The type of "s" has been "long" rather than the correct "unsigned long"
for quite some time.  This commit fixes this type error.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit abstracts funnel locking from synchronize_sched_expedited()
so that it may be used by synchronize_rcu_expedited().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Although synchronize_rcu_expedited() uses a sequence-counter scheme, it
is based on a single increment per grace period, which means that tasks
piggybacking off of concurrent grace periods may be forced to wait longer
than necessary.  This commit therefore applies the new sequence-count
functions developed for synchronize_sched_expedited() to speed things
up a bit and to consolidate the sequence-counter implementation.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit creates rcu_exp_gp_seq_start() and rcu_exp_gp_seq_end() to
bracket an expedited grace period, rcu_exp_gp_seq_snap() to snapshot the
sequence counter, and rcu_exp_gp_seq_done() to check to see if a full
expedited grace period has elapsed since the snapshot.  These will be
applied to synchronize_rcu_expedited().  These are defined in terms of
underlying rcu_seq_start(), rcu_seq_end(), rcu_seq_snap(), rcu_seq_done(),
which will be applied to _rcu_barrier().

One reason that this commit doesn't use the seqcount primitives themselves
is that the smp_wmb() in those primitive is insufficient due to the fact
that expedited grace periods do reads as well as writes.  In addition,
the read-side seqcount primitives detect a potentially partial change,
where the expedited primitives instead need a guaranteed full change.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Sequentially stopping the CPUs slows down expedited grace periods by
at least a factor of two, based on rcutorture's grace-period-per-second
rate.  This is a conservative measure because rcutorture uses unusually
long RCU read-side critical sections and because rcutorture periodically
quiesces the system in order to test RCU's ability to ramp down to and
up from the idle state.  This commit therefore replaces the stop_one_cpu()
with stop_one_cpu_nowait(), using an atomic-counter scheme to determine
when all CPUs have passed through the stopped state.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit gets rid of synchronize_sched_expedited()'s mutex_trylock()
polling loop in favor of a funnel-locking scheme based on the rcu_node
tree.  The work-done check is done at each level of the tree, allowing
high-contention situations to be resolved quickly with reasonable levels
of mutex contention.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Now that synchronize_sched_expedited() have a mutex, it can use simpler
work-already-done detection scheme.  This commit simplifies this scheme
by using something similar to the sequence-locking counter scheme.
A counter is incremented before and after each grace period, so that
the counter is odd in the midst of the grace period and even otherwise.
So if the counter has advanced to the second even number that is
greater than or equal to the snapshot, the required grace period has
already happened.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The synchronize_sched_expedited() currently invokes try_stop_cpus(),
which schedules the stopper kthreads on each online non-idle CPU,
and waits until all those kthreads are running before letting any
of them stop.  This is disastrous for real-time workloads, which
get hit with a preemption that is as long as the longest scheduling
latency on any CPU, including any non-realtime housekeeping CPUs.
This commit therefore switches to using stop_one_cpu() on each CPU
in turn.  This avoids inflicting the worst-case scheduling latency
on the worst-case CPU onto all other CPUs, and also simplifies the
code a little bit.

Follow-up commits will simplify the counter-snapshotting algorithm
and convert a number of the counters that are now protected by the
new ->expedited_mutex to non-atomic.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
[ paulmck: Kept stop_one_cpu(), dropped disabling of "guardrails". ]
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The CONFIG_RCU_CPU_STALL_INFO has been default-y for a couple of
releases with no complaints, so it is time to eliminate this Kconfig
option entirely, so that the long-form RCU CPU stall warnings cannot
be disabled.  This commit does just that.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

The fact that tasks could be migrated from leaf to root rcu_node
structures meant that synchronize_rcu_expedited() had to disable
CPU hotplug.  However, tasks now stay put, so this commit removes the
CPU-hotplug disabling from synchronize_rcu_expedited().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Currently if the rcu_fanout_leaf boot parameter is out of bounds (that
is, less than RCU_FANOUT_LEAF or greater than the number of bits in an
unsigned long), a warning is issued and execution continues with the
out-of-bounds value.  This can result in all manner of failures, so this
patch resets rcu_fanout_leaf to RCU_FANOUT_LEAF when an out-of-bounds
condition is detected.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Because gcc does not realize a loop would not be entered ever
(i.e. in case of rcu_num_lvls == 1):

  for (i = 1; i < rcu_num_lvls; i++)
	  rsp->level[i] = rsp->level[i - 1] + levelcnt[i - 1];

some compiler (pre- 5.x?) versions give a bogus warning:

  kernel/rcu/tree.c: In function ‘rcu_init_one.isra.55’:
  kernel/rcu/tree.c:4108:13: warning: array subscript is above array bounds [-Warray-bounds]
     rsp->level[i] = rsp->level[i - 1] + rsp->levelcnt[i - 1];
               ^
Fix that warning by adding an extra item to rcu_state::level[]
array. Once the bogus warning is fixed in gcc and kernel drops
support of older versions, the dummy item may be removed from
the array.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Suggested-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This update makes arithmetic to calculate number of RCU nodes
more straight and easy to read.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Although a number of RCU levels may be less than the current
maximum of four, some static data associated with each level
are allocated for all four levels. As result, the extra data
never get accessed and just wast memory. This update limits
count of allocated items to the number of used RCU levels.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Members rcu_state::levelcnt[] and rcu_state::levelspread[]
are only used at init. There is no reason to keep them
afterwards.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Number of items in rcu_capacity[] array is defined by macro
MAX_RCU_LVLS. However, that array is never accessed beyond
RCU_NUM_LVLS index. Therefore, we can limit the array to
RCU_NUM_LVLS items and eliminate MAX_RCU_LVLS. As result,
in most cases the memory is conserved.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Variable rcu_num_lvls is limited by RCU_NUM_LVLS macro.
In turn, rcu_state::levelcnt[] array is never accessed
beyond rcu_num_lvls. Thus, rcu_state::levelcnt[] is safe
to limit to RCU_NUM_LVLS items.

Since rcu_num_lvls could be changed during boot (as result
of rcutree.rcu_fanout_leaf kernel parameter update) one might
assume a new value could overflow the value of RCU_NUM_LVLS.
However, that is not the case, since leaf-level fanout is only
permitted to increase, resulting in rcu_num_lvls possibly to
decrease.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Current code suggests that introducing the extra level to
rcu_capacity[] array makes some of the arithmetic easier.
Well, in fact it appears rather confusing and unnecessary.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

This update simplifies rcu_init_geometry() code flow
and makes calculation of the total number of rcu_node
structures more easy to read.

The update relies on the fact num_rcu_lvl[] is never
accessed beyond rcu_num_lvls index by the rest of the
code. Therefore, there is no need initialize the whole
num_rcu_lvl[].

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Local variable 'n' mimics 'nr_cpu_ids' while the both are
used within one function. There is no reason for 'n' to
exist whatsoever.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Currently a condition when RCU tree is unable to accommodate
the configured number of CPUs is not permitted and causes
a fall back to compile-time values. However, the code has no
means to exceed the RCU tree capacity neither at compile-time
nor in run-time. Therefore, if the condition is met in run-
time then it indicates a serios problem elsewhere and should
be handled with a panic.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Merge tag 'platform-drivers-x86-v4.2-2' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86

Pull late x86 platform driver updates from Darren Hart:
 "The following came in a bit later and I wanted them to bake in next a
  few more days before submitting, thus the second pull.

  A new intel_pmc_ipc driver, a symmetrical allocation and free fix in
  dell-laptop, a couple minor fixes, and some updated documentation in
  the dell-laptop comments.

  intel_pmc_ipc:
   - Add Intel Apollo Lake PMC IPC driver

  tc1100-wmi:
   - Delete an unnecessary check before the function call "kfree"

  dell-laptop:
   - Fix allocating & freeing SMI buffer page
   - Show info about WiGig and UWB in debugfs
   - Update information about wireless control"

* tag 'platform-drivers-x86-v4.2-2' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86:
  intel_pmc_ipc: Add Intel Apollo Lake PMC IPC driver
  tc1100-wmi: Delete an unnecessary check before the function call "kfree"
  dell-laptop: Fix allocating & freeing SMI buffer page
  dell-laptop: Show info about WiGig and UWB in debugfs
  dell-laptop: Update information about wireless control

Pull more vfs updates from Al Viro:
 "Assorted VFS fixes and related cleanups (IMO the most interesting in
  that part are f_path-related things and Eric's descriptor-related
  stuff).  UFS regression fixes (it got broken last cycle).  9P fixes.
  fs-cache series, DAX patches, Jan's file_remove_suid() work"

[ I'd say this is much more than "fixes and related cleanups".  The
  file_table locking rule change by Eric Dumazet is a rather big and
  fundamental update even if the patch isn't huge.   - Linus ]

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (49 commits)
  9p: cope with bogus responses from server in p9_client_{read,write}
  p9_client_write(): avoid double p9_free_req()
  9p: forgetting to cancel request on interrupted zero-copy RPC
  dax: bdev_direct_access() may sleep
  block: Add support for DAX reads/writes to block devices
  dax: Use copy_from_iter_nocache
  dax: Add block size note to documentation
  fs/file.c: __fget() and dup2() atomicity rules
  fs/file.c: don't acquire files->file_lock in fd_install()
  fs:super:get_anon_bdev: fix race condition could cause dev exceed its upper limitation
  vfs: avoid creation of inode number 0 in get_next_ino
  namei: make set_root_rcu() return void
  make simple_positive() public
  ufs: use dir_pages instead of ufs_dir_pages()
  pagemap.h: move dir_pages() over there
  remove the pointless include of lglock.h
  fs: cleanup slight list_entry abuse
  xfs: Correctly lock inode when removing suid and file capabilities
  fs: Call security_ops->inode_killpriv on truncate
  fs: Provide function telling whether file_remove_privs() will do anything
  ...

Commit 835a6a2f ("Bluetooth: Stop sabotaging list poisoning")
thought that the code was sabotaging the list poisoning when NULL'ing
out the list pointers and removed it.

But what was going on was that the bluetooth code was using NULL
pointers for the list as a way to mark it empty, and that commit just
broke it (and replaced the test with NULL with a "list_empty()" test on
a uninitialized list instead, breaking things even further).

So fix it all up to use the regular and real list_empty() handling
(which does not use NULL, but a pointer to itself), also making sure to
initialize the list properly (the previous NULL case was initialized
implicitly by the session being allocated with kzalloc())

This is a combination of patches by Marcel Holtmann and Tedd Ho-Jeong
An.

[ I would normally expect to get this through the bt tree, but I'm going
  to release -rc1, so I'm just committing this directly   - Linus ]
Reported-and-tested-by: Jörg Otte <jrg.otte@gmail.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Original-by: Tedd Ho-Jeong An <tedd.an@intel.com>
Original-by: Marcel Holtmann <marcel@holtmann.org>:
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull SCSI target updates from Nicholas Bellinger:
 "It's been a busy development cycle for target-core in a number of
  different areas.

  The fabric API usage for se_node_acl allocation is now within
  target-core code, dropping the external API callers for all fabric
  drivers tree-wide.

  There is a new conversion to RCU hlists for se_node_acl and
  se_portal_group LUN mappings, that turns fast-past LUN lookup into a
  completely lockless code-path.  It also removes the original
  hard-coded limitation of 256 LUNs per fabric endpoint.

  The configfs attributes for backends can now be shared between core
  and driver code, allowing existing drivers to use common code while
  still allowing flexibility for new backend provided attributes.

  The highlights include:

   - Merge sbc_verify_dif_* into common code (sagi)
   - Remove iscsi-target support for obsolete IFMarker/OFMarker
     (Christophe Vu-Brugier)
   - Add bidi support in target/user backend (ilias + vangelis + agover)
   - Move se_node_acl allocation into target-core code (hch)
   - Add crc_t10dif_update common helper (akinobu + mkp)
   - Handle target-core odd SGL mapping for data transfer memory
     (akinobu)
   - Move transport ID handling into target-core (hch)
   - Move task tag into struct se_cmd + support 64-bit tags (bart)
   - Convert se_node_acl->device_list[] to RCU hlist (nab + hch +
     paulmck)
   - Convert se_portal_group->tpg_lun_list[] to RCU hlist (nab + hch +
     paulmck)
   - Simplify target backend driver registration (hch)
   - Consolidate + simplify target backend attribute implementations
     (hch + nab)
   - Subsume se_port + t10_alua_tg_pt_gp_member into se_lun (hch)
   - Drop lun_sep_lock for se_lun->lun_se_dev RCU usage (hch + nab)
   - Drop unnecessary core_tpg_register TFO parameter (nab)
   - Use 64-bit LUNs tree-wide (hannes)
   - Drop left-over TARGET_MAX_LUNS_PER_TRANSPORT limit (hannes)"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (76 commits)
  target: Bump core version to v5.0
  target: remove target_core_configfs.h
  target: remove unused TARGET_CORE_CONFIG_ROOT define
  target: consolidate version defines
  target: implement WRITE_SAME with UNMAP bit using ->execute_unmap
  target: simplify UNMAP handling
  target: replace se_cmd->execute_rw with a protocol_data field
  target/user: Fix inconsistent kmap_atomic/kunmap_atomic
  target: Send UA when changing LUN inventory
  target: Send UA upon LUN RESET tmr completion
  target: Send UA on ALUA target port group change
  target: Convert se_lun->lun_deve_lock to normal spinlock
  target: use 'se_dev_entry' when allocating UAs
  target: Remove 'ua_nacl' pointer from se_ua structure
  target_core_alua: Correct UA handling when switching states
  xen-scsiback: Fix compile warning for 64-bit LUN
  target: Remove TARGET_MAX_LUNS_PER_TRANSPORT
  target: use 64-bit LUNs
  target: Drop duplicate + unused se_dev_check_wce
  target: Drop unnecessary core_tpg_register TFO parameter
  ...

Pull NTB updates from Jon Mason:
 "This includes a pretty significant reworking of the NTB core code, but
  has already produced some significant performance improvements.

  An abstraction layer was added to allow the hardware and clients to be
  easily added.  This required rewriting the NTB transport layer for
  this abstraction layer.  This modification will allow future "high
  performance" NTB clients.

  In addition to this change, a number of performance modifications were
  added.  These changes include NUMA enablement, using CPU memcpy
  instead of asyncdma, and modification of NTB layer MTU size"

* tag 'ntb-4.2' of git://github.com/jonmason/ntb: (22 commits)
  NTB: Add split BAR output for debugfs stats
  NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe
  NTB: Print driver name and version in module init
  NTB: Increase transport MTU to 64k from 16k
  NTB: Rename Intel code names to platform names
  NTB: Default to CPU memcpy for performance
  NTB: Improve performance with write combining
  NTB: Use NUMA memory in Intel driver
  NTB: Use NUMA memory and DMA chan in transport
  NTB: Rate limit ntb_qp_link_work
  NTB: Add tool test client
  NTB: Add ping pong test client
  NTB: Add parameters for Intel SNB B2B addresses
  NTB: Reset transport QP link stats on down
  NTB: Do not advance transport RX on link down
  NTB: Differentiate transport link down messages
  NTB: Check the device ID to set errata flags
  NTB: Enable link for Intel root port mode in probe
  NTB: Read peer info from local SPAD in transport
  NTB: Split ntb_hw_intel and ntb_transport drivers
  ...

if server claims to have written/read more than we'd told it to,
warn and cap the claimed byte count to avoid advancing more than
we are ready to.

Braino in "9p: switch p9_client_write() to passing it struct iov_iter *";
if response is impossible to parse and we discard the request, get the
out of the loop right there.

Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

If we'd already sent a request and decide to abort it, we *must*
issue TFLUSH properly and not just blindly reuse the tag, or
we'll get seriously screwed when response eventually arrives
and we confuse it for response to later request that had reused
the same tag.

Cc: stable@vger.kernel.org # v3.2 and later
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

The brd driver is the only in-tree driver that may sleep currently.
After some discussion on linux-fsdevel, we decided that any driver
may choose to sleep in its ->direct_access method.  To ensure that all
callers of bdev_direct_access() are prepared for this, add a call
to might_sleep().
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

If a block device supports the ->direct_access methods, bypass the normal
DIO path and use DAX to go straight to memcpy() instead of allocating
a DIO and a BIO.

Includes support for the DIO_SKIP_DIO_COUNT flag in DAX, as is done in
do_blockdev_direct_IO().
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

When userspace does a write, there's no need for the written data to
pollute the CPU cache.  This matches the original XIP code.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

For block devices which are small enough, mkfs will default to creating
a filesystem with block sizes smaller than page size.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>