It has been exceptionally useful to exercise the logic that handles
local immediate errors and RDMA connection loss.  To enable
developers to test this regularly and repeatably, add logic to
simulate connection loss every so often.

Fault injection is disabled by default. It is enabled with

  $ sudo echo xxx > /sys/kernel/debug/sunrpc/inject_fault/disconnect

where "xxx" is a large positive number of transport method calls
before a disconnect. A value of several thousand is usually a good
number that allows reasonable forward progress while still causing a
lot of connection drops.

These hooks are disabled when SUNRPC_DEBUG is turned off.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

This was only ever set to nfs_writeback_release_common(), a function
which is completely empty.  Let's just drop this function pointer and
simplify the code a bit.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

nfs4_proc_lookup_common is supposed to return a posix error, we have to
handle any error returned that isn't errno
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Frank S. Filz <ffilzlnx@mindspring.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

RDMA xprts don't have a sock_xprt, but an rdma_xprt, so the
xs_swapper_enable/disable functions will likely oops when fed an RDMA
xprt. Turn these functions into rpc_xprt_ops so that that doesn't
occur. For now the RDMA versions are no-ops that just return -EINVAL
on an attempt to swapon.

Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

It's possible that we could race with a call to xs_reset_transport, in
which case the xprt->inet pointer could be zeroed out while we're
accessing it. Lock the xprt before we try to set memalloc on it.

Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

We currently increment the memalloc_socks counter if we have a xprt that
is associated with a swapfile. That socket can be replaced however
during a reconnect event, and the memalloc_socks counter is never
decremented if that occurs.

When tearing down a xprt socket, check to see if the xprt is set up for
swapping and sk_clear_memalloc before releasing the socket if so.
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Split xs_swapper into enable/disable functions and eliminate the
"enable" flag.

Currently, it's racy if you have multiple swapon/swapoff operations
running in parallel over the same xprt. Also fix it so that we only
set it to a memalloc socket on a 0->1 transition and only clear it
on a 1->0 transition.

Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Jerome reported seeing a warning pop when working with a swapfile on
NFS. The nfs_swap_activate can end up calling sk_set_memalloc while
holding the rcu_read_lock and that function can sleep.

To fix that, we need to take a reference to the xprt while holding the
rcu_read_lock, set the socket up for swapping and then drop that
reference. But, xprt_put is not exported and having NFS deal with the
underlying xprt is a bit of layering violation anyway.

Fix this by adding a set of activate/deactivate functions that take a
rpc_clnt pointer instead of an rpc_xprt, and have nfs_swap_activate and
nfs_swap_deactivate call those.

Also, add a per-rpc_clnt atomic counter to keep track of the number of
active swapfiles associated with it. When the counter does a 0->1
transition, we enable swapping on the xprt, when we do a 1->0 transition
we disable swapping on it.

This also allows us to be a bit more selective with the RPC_TASK_SWAPPER
flag. If non-swapper and swapper clnts are sharing a xprt, then we only
need to flag the tasks from the swapper clnt with that flag.
Acked-by: Mel Gorman <mgorman@suse.de>
Reported-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

We need to allow the server to send a new request immediately after we've
replied to the previous one. Right now, there is a window between the
send and the release of the old request in rpc_put_task(), where the
server could send us a new backchannel RPC call, and we have no
request to service it.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Clean up: Merge bc_send() into bc_svc_process().

Note: even thought this touches svc.c, it is a client-side change.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

If the socket was busy due to a socket nospace error, then we should
retry the send.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

req->rq_private_buf isn't initialised when xprt_setup_backchannel calls
xprt_free_allocation.

Fixes: fb7a0b9a ("nfs41: New backchannel helper routines")
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Delete jump to a label on the next line, when that label is not
used elsewhere.

A simplified version of the semantic patch that makes this change is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@r@
identifier l;
@@

-if (...) goto l;
-l:
// </smpl>

Also drop the unnecessary ret variable.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Several functions have outdated arguments listed in the doc comments.
Drop documentation for arguments that no longer exist.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

When encoding the NFSACL SETACL operation, reserve just the estimated
size of the ACL rather than a fixed maximum. This eliminates needless
zero padding on the wire that the server ignores.

Fixes: ee5dc773 ('NFS: Fix "kernel BUG at fs/nfs/nfs3xdr.c:1338!"')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

In glibc 2.21 (and several previous), a call to opendir() will
result in a 32K (BUFSIZ*4) buffer being allocated and passed to
getdents.

However a call to fdopendir() results in an 'fstat' request to
determine block size and a matching buffer allocated for subsequent
use with getdents.  This will typically be 1M.

The first getdents call on an NFS directory will always use
READDIR_PLUS (or NFSv4 equivalent) if available.  Subsequent getdents
calls only use this more expensive version if some 'stat' requests are
made between the getdents calls.

For this reason it is good to keep at least that first getdents call
relatively short.  When fdopendir() and readdir() is used on a large
directory, it takes approximately 32 times as long to complete as
using "opendir".  Current versions of 'find' use fdopendir() and
demonstrate this slowness.

'stat' on a directory currently returns the 'wsize'.  This number has
no meaning on directories.
Actual READDIR requests are limited to ->dtsize, which itself is
capped at 4 pages, coincidently the same as BUFSIZ*4.
So this is a meaningful number to use as the blocksize on directories,
and has the effect of making 'find' on large directories go a lot
faster.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

While the NFSv4.1 code has always drained the slot tables in order to stop
non-recovery related RPC calls when doing lease recovery, the NFSv4 code
did not.
The reason for the difference in behaviour is that NFSv4 does not have
session state, and so RPC calls can in theory proceed while recovery is
happening. In practice, however, anything I/O or state related needs to
wait until recovery is over.

This patch changes the behaviour of NFSv4 to match that of NFSv4.1 so that
we can simplify the state recovery code by assuming that we do not have to
deal with races between recovery and ordinary I/O.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Problem: When an operation like WRITE receives a BAD_STATEID, even though
recovery code clears the RECLAIM_NOGRACE recovery flag before recovering
the open state, because of clearing delegation state for the associated
inode, nfs_inode_find_state_and_recover() gets called and it makes the
same state with RECLAIM_NOGRACE flag again. As a results, when we restart
looking over the open states, we end up in the infinite loop instead of
breaking out in the next test of state flags.

Solution: unset the RECLAIM_NOGRACE set because of
calling of nfs_inode_find_state_and_recover() after returning from calling
recover_open() function.
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

Pull vfs fix from Al Viro:
 "Off-by-one in d_walk()/__dentry_kill() race fix.

  It's very hard to hit; possible in the same conditions as the original
  bug, except that you need the skipped branch to contain all the
  remaining evictables, so that the d_walk()-calling loop in
  d_invalidate() decides there's nothing more to do and doesn't go for
  another pass - otherwise that next pass will sweep the sucker.

  So it's not too urgent, but seeing that the fix is obvious and the
  original commit has spread into all -stable branches..."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  d_walk() might skip too much

Pull ARM fixes from Russell King:
 "Three fixes this time around:

   - fix a memory leak which occurs when probing performance monitoring
     unit interrupts

   - fix handling of non-PMD aligned end of RAM causing boot failures

   - fix missing syscall trace exit path with syscall tracing enabled
     causing a kernel oops in the audit code"

* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
  ARM: 8357/1: perf: fix memory leak when probing PMU PPIs
  ARM: fix missing syscall trace exit
  ARM: 8356/1: mm: handle non-pmd-aligned end of RAM

Pull MIPS fixes from Ralf Baechle:
 "MIPS fixes for 4.1 all across the tree"

* 'upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/ralf/linux:
  MIPS: strnlen_user.S: Fix a CPU_DADDI_WORKAROUNDS regression
  MIPS: BMIPS: Fix bmips_wr_vec()
  MIPS: ath79: fix build problem if CONFIG_BLK_DEV_INITRD is not set
  MIPS: Fuloong 2E: Replace CONFIG_USB_ISP1760_HCD by CONFIG_USB_ISP1760
  MIPS: irq: Use DECLARE_BITMAP
  ttyFDC: Fix to use native endian MMIO reads
  MIPS: Fix CDMM to use native endian MMIO reads

Pull turbostat tool fixes from Len Brown:
 "Just one minor kernel dependency in this batch -- added a #define to
  msr-index.h"

* 'turbostat' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux:
  tools/power turbostat: update version number to 4.7
  tools/power turbostat: allow running without cpu0
  tools/power turbostat: correctly decode of ENERGY_PERFORMANCE_BIAS
  tools/power turbostat: enable turbostat to support Knights Landing (KNL)
  tools/power turbostat: correctly display more than 2 threads/core

Pull SCSI target fixes from Nicholas Bellinger:
 "These are mostly minor fixes, with the exception of the following that
  address fall-out from recent v4.1-rc1 changes:

   - regression fix related to the big fabric API registration changes
     and configfs_depend_item() usage, that required cherry-picking one
     of HCH's patches from for-next to address the issue for v4.1 code.

   - remaining TCM-USER -v2 related changes to enforce full CDB
     passthrough from Andy + Ilias.

  Also included is a target_core_pscsi driver fix from Andy that
  addresses a long standing issue with a Scsi_Host reference being
  leaked on PSCSI device shutdown"

* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
  iser-target: Fix error path in isert_create_pi_ctx()
  target: Use a PASSTHROUGH flag instead of transport_types
  target: Move passthrough CDB parsing into a common function
  target/user: Only support full command pass-through
  target/user: Update example code for new ABI requirements
  target/pscsi: Don't leak scsi_host if hba is VIRTUAL_HOST
  target: Fix se_tpg_tfo->tf_subsys regression + remove tf_subsystem
  target: Drop signal_pending checks after interruptible lock acquire
  target: Add missing parentheses
  target: Fix bidi command handling
  target/user: Disallow full passthrough (pass_level=0)
  ISCSI: fix minor memory leak

Merge tag 'hwmon-for-linus-v4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging

Pull hwmon fixes from Guenter Roeck:
 "Some late hwmon patches, all headed for -stable

   - fix sysfs attribute initialization in nct6775 and nct6683 drivers

   - do not attempt to auto-detect tmp435 on I2C address 0x37

   - ensure iio channel is of type IIO_VOLTAGE in ntc_thermistor driver"

* tag 'hwmon-for-linus-v4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (nct6683) Add missing sysfs attribute initialization
  hwmon: (nct6775) Add missing sysfs attribute initialization
  hwmon: (tmp401) Do not auto-detect chip on I2C address 0x37
  hwmon: (ntc_thermistor) Ensure iio channel is of type IIO_VOLTAGE

We don't assign pi_ctx to desc->pi_ctx until we're certain to succeed
in the function.  That means the cleanup path should use the local
pi_ctx variable, not desc->pi_ctx.

This was detected by Coverity (CID 1260062).
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

It seems like we only care if a transport is passthrough or not. Convert
transport_type to a flags field and replace TRANSPORT_PLUGIN_* with a
flag, TRANSPORT_FLAG_PASSTHROUGH.
Signed-off-by: Andy Grover <agrover@redhat.com>
Reviewed-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

Aside from whether they handle BIDI ops or not, parsing of the CDB by
kernel and user SCSI passthrough modules should be identical. Move this
into a new passthrough_parse_cdb() and call it from tcm-pscsi and tcm-user.
Reported-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

After much discussion, give up on only passing a subset of SCSI commands
to userspace and pass them all. Based on what pscsi is doing, make sure
to set SCF_SCSI_DATA_CDB for I/O ops, and define attributes identical to
pscsi.

Make hw_block_size configurable via dev param.

Remove mention of command filtering from tcmu-design.txt.
Signed-off-by: Andy Grover <agrover@redhat.com>
Reviewed-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

We now require that the userspace handler set a bit if the command is not
handled.

Update calls to tcmu_hdr_get_op for v2.
Signed-off-by: Andy Grover <agrover@redhat.com>
Reviewed-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

See https://bugzilla.redhat.com/show_bug.cgi?id=1025672

We need to put() the reference to the scsi host that we got in
pscsi_configure_device(). In VIRTUAL_HOST mode it is associated with
the dev_virt, not the hba_virt.
Signed-off-by: Andy Grover <agrover@redhat.com>
Cc: stable@vger.kernel.org # 2.6.38+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

There is just one configfs subsystem in the target code, so we might as
well add two helpers to reference / unreference it from the core code
instead of passing pointers to it around.

This fixes a regression introduced for v4.1-rc1 with commit 9ac8928e,
where configfs_depend_item() callers using se_tpg_tfo->tf_subsys would
fail, because the assignment from the original target_core_subsystem[]
is no longer happening at target_register_template() time.

(Fix target_core_exit_configfs pointer dereference - Sagi)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reported-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

The following error message is seen when loading the nct6683 driver
with DEBUG_LOCK_ALLOC enabled.

BUG: key ffff88040b2f0030 not in .data!
------------[ cut here ]------------
WARNING: CPU: 0 PID: 186 at kernel/locking/lockdep.c:2988
				lockdep_init_map+0x469/0x630()
DEBUG_LOCKS_WARN_ON(1)

Caused by a missing call to sysfs_attr_init() when initializing
sysfs attributes.
Reported-by: Alexey Orishko <alexey.orishko@gmail.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.18+
Signed-off-by: Guenter Roeck <linux@roeck-us.net>

The following error message is seen when loading the nct6775 driver
with DEBUG_LOCK_ALLOC enabled.

BUG: key ffff88040b2f0030 not in .data!
------------[ cut here ]------------
WARNING: CPU: 0 PID: 186 at kernel/locking/lockdep.c:2988
				lockdep_init_map+0x469/0x630()
DEBUG_LOCKS_WARN_ON(1)

Caused by a missing call to sysfs_attr_init() when initializing
sysfs attributes.
Reported-by: Alexey Orishko <alexey.orishko@gmail.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.12+
Signed-off-by: Guenter Roeck <linux@roeck-us.net>

Pull PCI / ACPI fix from Rafael Wysocki:
 "This fixes a bug uncovered by a recent driver core change that
  modified the implementation of the ACPI_COMPANION_SET() macro to
  strictly rely on its second argument to be either NULL or a valid
  pointer to struct acpi_device.

  As it turns out, pcibios_root_bridge_prepare() on x86 and ia64 works
  with the assumption that the only code path calling pci_create_root_bus()
  is pci_acpi_scan_root() and therefore the sysdata argument passed to
  it will always match the expectations of pcibios_root_bridge_prepare().

  That need not be the case, however, and in particular it is not the
  case for the Xen pcifront driver that passes a pointer to its own
  private data strcture as sysdata to pci_scan_bus_parented() which then
  passes it to pci_create_root_bus() and it ends up being used incorrectly
  by pcibios_root_bridge_prepare()"

* tag 'acpi-pci-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  PCI / ACPI: Do not set ACPI companions for host bridges with parents

Pull xfs fixes from Dave Chinner:
 "This is a little larger than I'd like late in the release cycle, but
  all the fixes are for regressions introduced in the 4.1-rc1 merge, or
  are needed back in -stable kernels fairly quickly as they are
  filesystem corruption or userspace visible correctness issues.

  Changes in this update:

   - regression fix for new rename whiteout code

   - regression fixes for new superblock generic per-cpu counter code

   - fix for incorrect error return sign introduced in 3.17

   - metadata corruption fixes that need to go back to -stable kernels"

* tag 'xfs-for-linus-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs:
  xfs: fix broken i_nlink accounting for whiteout tmpfile inode
  xfs: xfs_iozero can return positive errno
  xfs: xfs_attr_inactive leaves inconsistent attr fork state behind
  xfs: extent size hints can round up extents past MAXEXTLEN
  xfs: inode and free block counters need to use __percpu_counter_compare
  percpu_counter: batch size aware __percpu_counter_compare()
  xfs: use percpu_counter_read_positive for mp->m_icount

Pull ARM SoC fixes from Arnd Bergmann:
 "Two weeks worth of small bug fixes this time, nothing sticking out
  this time:

   - one defconfig change to adapt to a modified Kconfig symbol

   - two fixes for i.MX for backwards compatibility with older DT files
     that was accidentally broken

   - one regression fix for irq handling on pxa

   - three small dt files on omap, and one each for imx and exynos"

* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  ARM: multi_v7_defconfig: Replace CONFIG_USB_ISP1760_HCD by CONFIG_USB_ISP1760
  ARM: imx6: gpc: don't register power domain if DT data is missing
  ARM: imx6: allow booting with old DT
  ARM: dts: set display clock correctly for exynos4412-trats2
  ARM: pxa: pxa_cplds: signedness bug in probe
  ARM: dts: Fix WLAN interrupt line for AM335x EVM-SK
  ARM: dts: omap3-devkit8000: Fix NAND DT node
  ARM: dts: am335x-boneblack: disable RTC-only sleep
  ARM: dts: fix imx27 dtb build rule
  ARM: dts: imx27: only map 4 Kbyte for fec registers

Pull device-mapper fixes from Mike Snitzer:
 "Quite a few fixes for DM's blk-mq support thanks to extra DM multipath
  testing from Junichi Nomura and Bart Van Assche.

  Also fix a casting bug in dm_merge_bvec() that could cause only a
  single page to be added to a bio (Joe identified this while testing
  dm-cache writeback)"

* tag 'dm-4.1-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: fix casting bug in dm_merge_bvec()
  dm: fix reload failure of 0 path multipath mapping on blk-mq devices
  dm: fix false warning in free_rq_clone() for unmapped requests
  dm: requeue from blk-mq dm_mq_queue_rq() using BLK_MQ_RQ_QUEUE_BUSY
  dm mpath: fix leak of dm_mpath_io structure in blk-mq .queue_rq error path
  dm: fix NULL pointer when clone_and_map_rq returns !DM_MAPIO_REMAPPED
  dm: run queue on re-queue