openwrt/staging/blogic.git
8 years agoqla2xxx: Add selective command queuing
Quinn Tran [Thu, 17 Dec 2015 19:57:06 +0000 (14:57 -0500)]
qla2xxx: Add selective command queuing

queue work element to specific process lessen cache miss

Signed-off-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Add irq affinity notification
Quinn Tran [Thu, 17 Dec 2015 19:57:05 +0000 (14:57 -0500)]
qla2xxx: Add irq affinity notification

Register to receive notification of when irq setting change
occured.

Signed-off-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Remove dependency on hardware_lock to reduce lock contention.
Quinn Tran [Thu, 17 Dec 2015 19:57:04 +0000 (14:57 -0500)]
qla2xxx: Remove dependency on hardware_lock to reduce lock contention.

Sessions management (add, deleted, modify) currently are serialized
through the hardware_lock.  Hardware_lock is a high traffic lock.
This lock is accessed by both the transmit & receive sides.

Sessions management is now moved off to another lock call sess_lock.
This is done to reduce lock contention and increase traffic throughput.

Signed-off-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Replace QLA_TGT_STATE_ABORTED with a bit.
Quinn Tran [Thu, 17 Dec 2015 19:57:03 +0000 (14:57 -0500)]
qla2xxx: Replace QLA_TGT_STATE_ABORTED with a bit.

Replace QLA_TGT_STATE_ABORTED state with a bit because
the current state of the command is lost when an abort
is requested by upper layer.

Signed-off-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Wait for all conflicts before ack'ing PLOGI
Alexei Potashnik [Thu, 17 Dec 2015 19:57:02 +0000 (14:57 -0500)]
qla2xxx: Wait for all conflicts before ack'ing PLOGI

Until now ack'ing of a new PLOGI has only been delayed if there
was an existing session for the same WWN. Ack was released when
the session deletion completed.

If there was another WWN session with the same fc_id/loop_id pair
(aka "conflicting session"), PLOGI was still ack'ed immediately.
This potentially caused a problem when old session deletion logged
fc_id/loop_id out of FW after new session has been established.

Two work-arounds were attempted before:
1. Dropping PLOGIs until conflicting session goes away.
2. Detecting initiator being logged out of FW and issuing LOGO
to force re-login.

This patch introduces proper solution to the problem where PLOGI
is held until either existing session with same WWN or any
conflicting session goes away. Mechanism supports one session holding
two PLOGI acks as well as one PLOGI ack being held by many sessions.

Signed-off-by: Alexei Potashnik <alexei@purestorage.com>
Acked-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Delete session if initiator is gone from FW
Alexei Potashnik [Thu, 17 Dec 2015 19:57:01 +0000 (14:57 -0500)]
qla2xxx: Delete session if initiator is gone from FW

1. Initiator A is logged in with fc_id(1)/loop_id(1)
2. Initiator A re-logs in with fc_id(2)/loop_id(2)
3. Part of old session deletion async logoout for 1/1 is queued
4. Initiator B logs in with fc_id(1)/loop_id(1), starts
   passing data and creates session.
5. Async logo from 3 is processed by DPC and sent to FW

Now initiator B has the session but is logged out from FW.

This condition is detected first with CTIO error 29 at which
point we should delete current session. During session
deletion we will send LOGO to initiator to force re-login.

Under rare circumstances initiator might be logged out of FW,
not have driver session, but still think it's logged in.
E.g. the above sequence plus session deletion due to re-config.
Incoming commands will fail to create local session because
initiator is not found in FW. In this case we also issue LOGO
to initiator to force him re-login.

Finally this patch fixes exchange leak when commands where
received in logged out state. In this case loop_id must be
set to FFFF when corresponding exchange is terminated. The
patch modifies exchange termination to always use FFFF,
since in certain scenarios it's impossible to tell whether
command was received in logged in or logged out state.

Signed-off-by: Alexei Potashnik <alexei@purestorage.com>
Acked-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Added interface to send explicit LOGO.
Himanshu Madhani [Thu, 17 Dec 2015 19:57:00 +0000 (14:57 -0500)]
qla2xxx: Added interface to send explicit LOGO.

This patch adds interface to send explicit LOGO
explicit LOGO using using ELS commands from driver.

Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Add FW resource count in DebugFS.
Quinn Tran [Thu, 17 Dec 2015 19:56:59 +0000 (14:56 -0500)]
qla2xxx: Add FW resource count in DebugFS.

DebugFS now will show fw_resource_count node.

FW Resource count

Original TGT exchg count[0]
current TGT exchg count[0]
original Initiator Exchange count[2048]
Current Initiator Exchange count[2048]
Original IOCB count[2078]
Current IOCB count[2067]
MAX VP count[254]
MAX FCF count[0]

Signed-off-by: Quinn Tran <quinn.tran@qlogic.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Enable Target counters in DebugFS.
Himanshu Madhani [Thu, 17 Dec 2015 19:56:58 +0000 (14:56 -0500)]
qla2xxx: Enable Target counters in DebugFS.

Following counters are added in target mode to help debugging efforts.

Target Counters

qla_core_sbt_cmd = 0
qla_core_ret_sta_ctio = 0
qla_core_ret_ctio = 0
core_qla_que_buf = 0
core_qla_snd_status = 0
core_qla_free_cmd = 0
num alloc iocb failed = 0
num term exchange sent = 0
num Q full sent = 0

Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Enable Exchange offload support.
Himanshu Madhani [Thu, 17 Dec 2015 19:56:57 +0000 (14:56 -0500)]
qla2xxx: Enable Exchange offload support.

This patch enables Exchange offload support in Qlogic ISP.
To enable exchange offload with Qlogic ISP24XX/25XX/26XX,
set module parameter ql2xexchoffld to any non-zero number.
This will alow ISP firmware to store exchange data structures
used by firmware to host memory provided by driver. ISP firmware
can supports upto 32k total active exchanges.

Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoqla2xxx: Enable Extended Logins support
Himanshu Madhani [Thu, 17 Dec 2015 19:56:56 +0000 (14:56 -0500)]
qla2xxx: Enable Extended Logins support

This patch enables Extended Logins support in Qlogic ISP.
To enable extended login with Qlogic ISP24XX/25XX/26XX,
set module parameter ql2xexlogins to any non-zero number.
This will alow ISP firmware to store port database structure
information of remote login sessions to host memory provided
by driver. ISP firmware supports upto 16k total logins.

Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotarget: use offset_in_page macro
Geliang Tang [Wed, 25 Nov 2015 13:49:27 +0000 (21:49 +0800)]
target: use offset_in_page macro

Use offset_in_page macro instead of (addr & ~PAGE_MASK).

Signed-off-by: Geliang Tang <geliangtang@163.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotarget: fix deprecated attribute names in dmesg
Christophe Vu-Brugier [Wed, 18 Nov 2015 08:22:58 +0000 (09:22 +0100)]
target: fix deprecated attribute names in dmesg

The following message is displayed in dmesg when a deprecated attribute
is set:

  "ignoring deprecated ##_name## attribute"

This patch fixes the format to include the name of the deprecated
attribute.

Signed-off-by: Christophe Vu-Brugier <cvubrugier@fastmail.fm>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotarget/sbc: Add LBPRZ attribute + control CDB emulation
Jamie Pocas [Sun, 29 Nov 2015 22:44:57 +0000 (14:44 -0800)]
target/sbc: Add LBPRZ attribute + control CDB emulation

This change sets the LBPRZ flag in EVPD page b2h and READ CAPACITY (16)
based on a new unmap_zeroes_data device attribute. This flag is set
automatically for iblock based on underlying block device queue's
discard_zeroes_data flag.

Signed-off-by: Jamie Pocas <jamie.pocas@emc.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agosbp-target: Remove a superfluous forward declaration
Bart Van Assche [Thu, 22 Oct 2015 22:56:25 +0000 (15:56 -0700)]
sbp-target: Remove a superfluous forward declaration

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Acked-by: Chris Boot <bootc@bootc.net>
Reviewed-by: Andy Grover <agrover@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agoiscsi-target: Fix indentation + spelling + unreachable code
Bart Van Assche [Thu, 22 Oct 2015 22:53:22 +0000 (15:53 -0700)]
iscsi-target: Fix indentation + spelling + unreachable code

Modify indentation such that the 'smatch' tool no longer complains
about incorrect indentation + unreachable code.

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotarget: Fix spelling + remove set-but-not-used variables
Bart Van Assche [Thu, 22 Oct 2015 22:52:01 +0000 (15:52 -0700)]
target: Fix spelling + remove set-but-not-used variables

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Andy Grover <agrover@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: f_tcm: add configfs support
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:26 +0000 (16:06 +0100)]
usb: gadget: f_tcm: add configfs support

Allow using the tcm function as a component of a gadget composed with
ConfigFS.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: f_tcm: use usb_gstrings_attach
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:25 +0000 (16:06 +0100)]
usb: gadget: f_tcm: use usb_gstrings_attach

Do not directly use file static strings definitions in instances of f_tcm.
Instead use usb_gstrings_attach.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: f_tcm: remove redundant singleton
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:24 +0000 (16:06 +0100)]
usb: gadget: f_tcm: remove redundant singleton

The only instance is guaranteed with TPG_INSTANCES defined to 1.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: f_tcm: remove compatibility layer
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:23 +0000 (16:06 +0100)]
usb: gadget: f_tcm: remove compatibility layer

There are no old function interface users left.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: convert to use new function registration interface
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:22 +0000 (16:06 +0100)]
usb: gadget: tcm: convert to use new function registration interface

Convert the only user of old tcm function interface so that the old
interface can be removed.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: f_tcm: convert to new function interface with backward compatibility
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:21 +0000 (16:06 +0100)]
usb: gadget: f_tcm: convert to new function interface with backward compatibility

Converting tcm to the new function interface requires converting
USB tcm's function code and its users.

This patch converts the f_tcm.c to the new function interface.

The file can be now compiled into a separate module usb_f_tcm.ko.

The old function interface is provided by means of preprocessor conditional
directives. After all users are converted, the old interface can be
removed.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: factor out f_tcm
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:20 +0000 (16:06 +0100)]
usb: gadget: tcm: factor out f_tcm

Prepare for converting tcm to new function registration interface.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: simplify attribute store function
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:19 +0000 (16:06 +0100)]
usb: gadget: tcm: simplify attribute store function

Simplify function code.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: use strtobool for a boolean value
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:18 +0000 (16:06 +0100)]
usb: gadget: tcm: use strtobool for a boolean value

Simplify the function.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: follow naming conventions
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:17 +0000 (16:06 +0100)]
usb: gadget: tcm: follow naming conventions

Prepare for splitting tcm_usb_gadget into legacy gadget proper and f_tcm.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agousb: gadget: tcm: split string definitions into function and device
Andrzej Pietrasiewicz [Fri, 11 Dec 2015 15:06:16 +0000 (16:06 +0100)]
usb: gadget: tcm: split string definitions into function and device

Prepare for factoring out f_tcm from a legacy gadget.

Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotcm_usb_gadget: Fix enabled attribute failure
Nicholas Bellinger [Fri, 11 Dec 2015 15:06:15 +0000 (16:06 +0100)]
tcm_usb_gadget: Fix enabled attribute failure

Fix up tcm_usbg_tpg_store_enable() return value to propagate
usbg_attach() failure up to user-space if no HDC is found.

Reported-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotcm_usb_gadget: Fix nexus leak
Nicholas Bellinger [Fri, 11 Dec 2015 15:06:14 +0000 (16:06 +0100)]
tcm_usb_gadget: Fix nexus leak

This patch adds the missing tcm_usbg_drop_nexus() to properly
release tcm_usbg_nexus memory during typical ->fabric_drop_tpg()
callback shutdown.

Reported-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotcm_usb_gadget: Don't strip off nexus WWPN prefix
Nicholas Bellinger [Fri, 11 Dec 2015 15:06:13 +0000 (16:06 +0100)]
tcm_usb_gadget: Don't strip off nexus WWPN prefix

Avoid stripping off the 'naa.' I_T nexus prefix from configfs
attribute store input, so that user-space will get back what
it originaly wrote into ../usb_gadget/$WWPN/$TPGT/nexus.

Note the SCSI initiator WWPN is purely symbolic for UAS + BOT,
so it will not effect host side code.

Reported-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agofs: configfs: Add unlocked version of configfs_depend_item()
Krzysztof Opasiak [Fri, 11 Dec 2015 15:06:12 +0000 (16:06 +0100)]
fs: configfs: Add unlocked version of configfs_depend_item()

This change is necessary for the SCSI target usb gadget composed with
configfs. In this case configfs will be used for two different purposes:
to compose a usb gadget and to configure the target part. If an instance
of tcm function is created in $CONFIGFS_ROOT/usb_gadget/<gadget>/functions
a tpg can be created in $CONFIGFS_ROOT/target/usb_gadget/<wwn>/, but after
a tpg is created the tcm function must not be removed until its
corresponding tpg is gone. While the configfs_depend/undepend_item() are
meant exactly for creating this kind of dependencies, they are not suitable
if the other kernel subsystem happens to be another subsystem in configfs,
so this patch adds unlocked versions meant for configfs callbacks.

Above description has been provided by:
Andrzej Pietrasiewicz <andrzej.p@samsung.com>

In configfs_depend_item() we have to consider two possible cases:

1) When we are called to depend another item in the same subsystem
   as caller
In this case we should skip locking configfs root as we know
that configfs is in valid state and our subsystem will not
be unregistered during this call.

2) When we are called to depend item in different subsystem than
   our caller
In this case we are also sure that configfs is in valid state
but we have to lock root of configfs to avoid unregistration
of target's subsystem. As it is other than caller's subsystem,
there may be nothing what protects us against unregistration
of that subsystem.

Signed-off-by: Krzysztof Opasiak <k.opasiak@samsung.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agofs: configfs: Factor out configfs_find_subsys_dentry()
Krzysztof Opasiak [Fri, 11 Dec 2015 15:06:11 +0000 (16:06 +0100)]
fs: configfs: Factor out configfs_find_subsys_dentry()

configfs_depend_item() is quite complicated and should
be split up into smaller functions. This also allow to
share this code with other functions.

Signed-off-by: Krzysztof Opasiak <k.opasiak@samsung.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agofs: configfs: Factor out configfs_do_depend_item()
Krzysztof Opasiak [Fri, 11 Dec 2015 15:06:10 +0000 (16:06 +0100)]
fs: configfs: Factor out configfs_do_depend_item()

configfs_depend_item() is quite complicated and should
be split up into smaller functions. This also allow to
share this code with other functions.

Signed-off-by: Krzysztof Opasiak <k.opasiak@samsung.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agofs: configfs: Drop unused parameter from configfs_undepend_item()
Krzysztof Opasiak [Fri, 11 Dec 2015 15:06:09 +0000 (16:06 +0100)]
fs: configfs: Drop unused parameter from configfs_undepend_item()

subsys parameter is never used by configfs_undepend_item()
so there is no point in passing it to this function.

Signed-off-by: Krzysztof Opasiak <k.opasiak@samsung.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
8 years agotcm_loop: Show address of tpg in configfs
Sheng Yang [Sat, 14 Nov 2015 22:17:16 +0000 (14:17 -0800)]
tcm_loop: Show address of tpg in configfs

In the past, the scsi_host's number wasn't shown anywhere, user have
to depends on vpg_83 to find the block device, which is also depends
on backstore implementation.

It's better for tcm_loop to provide necessary straightforward
information on locate the block device it created.

This patch would help to locate the block device created by tcm_loop.
The address would be shown at e.g.

  /sys/kernel/config/target/loopback/naa.60014059436855c1/tpgt_1/address

which would looks like "2:0:1", and the lun number can be found at
"/tpgt_1/lun/lun_0".  Altogether they formated the scsi address of
device as "2:0:1:0", which can be used to locate the device easily
through 'lsscsi'.

(Update to >= v4.4-rc1 configfs attribute usage - nab)

Signed-off-by: Sheng Yang <sheng@yasker.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
9 years agoLinux 4.4-rc2
Linus Torvalds [Mon, 23 Nov 2015 00:45:59 +0000 (16:45 -0800)]
Linux 4.4-rc2

9 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Sun, 22 Nov 2015 23:21:40 +0000 (15:21 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge slub bulk allocator updates from Andrew Morton:
 "This missed the merge window because I was waiting for some repairs to
  come in.  Nothing actually uses the bulk allocator yet and the changes
  to other code paths are pretty small.  And the net guys are waiting
  for this so they can start merging the client code"

More comments from Jesper Dangaard Brouer:
 "The kmem_cache_alloc_bulk() call, in mm/slub.c, were included in
  previous kernel.  The present version contains a bug.  Vladimir
  Davydov noticed it contained a bug, when kernel is compiled with
  CONFIG_MEMCG_KMEM (see commit 03ec0ed57ffc: "slub: fix kmem cgroup
  bug in kmem_cache_alloc_bulk").  Plus the mem cgroup counterpart in
  kmem_cache_free_bulk() were missing (see commit 033745189b1b "slub:
  add missing kmem cgroup support to kmem_cache_free_bulk").

  I don't consider the fix stable-material because there are no in-tree
  users of the API.

  But with known bugs (for memcg) I cannot start using the API in the
  net-tree"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  slab/slub: adjust kmem_cache_alloc_bulk API
  slub: add missing kmem cgroup support to kmem_cache_free_bulk
  slub: fix kmem cgroup bug in kmem_cache_alloc_bulk
  slub: optimize bulk slowpath free by detached freelist
  slub: support for bulk free with SLUB freelists

9 years agoMerge tag 'tty-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Linus Torvalds [Sun, 22 Nov 2015 23:10:57 +0000 (15:10 -0800)]
Merge tag 'tty-4.4-rc2' of git://git./linux/kernel/git/gregkh/tty

Pull tty/serial fixes from Greg KH:
 "Here are a few small tty/serial driver fixes for 4.4-rc2 that resolve
  some reported problems.

  All have been in linux-next, full details are in the shortlog below"

* tag 'tty-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: export fsl8250_handle_irq
  serial: 8250_mid: Add missing dependency
  tty: audit: Fix audit source
  serial: etraxfs-uart: Fix crash
  serial: fsl_lpuart: Fix earlycon support
  bcm63xx_uart: Use the device name when registering an interrupt
  tty: Fix direct use of tty buffer work
  tty: Fix tty_send_xchar() lock order inversion

9 years agoMerge tag 'staging-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sun, 22 Nov 2015 21:26:24 +0000 (13:26 -0800)]
Merge tag 'staging-4.4-rc2' of git://git./linux/kernel/git/gregkh/staging

Pull staging/IIO fixes from Greg KH:
 "Here are some staging and iio driver fixes for 4.4-rc2.  All of these
  are in response to issues that have been reported and have been in
  linux-next for a while"

* tag 'staging-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  Revert "Staging: wilc1000: coreconfigurator: Drop unneeded wrapper functions"
  iio: adc: xilinx: Fix VREFN scale
  iio: si7020: Swap data byte order
  iio: adc: vf610_adc: Fix division by zero error
  iio:ad7793: Fix ad7785 product ID
  iio: ad5064: Fix ad5629/ad5669 shift
  iio:ad5064: Make sure ad5064_i2c_write() returns 0 on success
  iio: lpc32xx_adc: fix warnings caused by enabling unprepared clock
  staging: iio: select IRQ_WORK for IIO_DUMMY_EVGEN
  vf610_adc: Fix internal temperature calculation

9 years agoMerge tag 'usb-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Sun, 22 Nov 2015 21:15:05 +0000 (13:15 -0800)]
Merge tag 'usb-4.4-rc2' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are a number of USB fixes and new device ids for 4.4-rc2.  All
  have been in linux-next and the details are in the shortlog"

* tag 'usb-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (28 commits)
  usblp: do not set TASK_INTERRUPTIBLE before lock
  USB: MAINTAINERS: cxacru
  usb: kconfig: fix warning of select USB_OTG
  USB: option: add XS Stick W100-2 from 4G Systems
  xhci: Fix a race in usb2 LPM resume, blocking U3 for usb2 devices
  usb: xhci: fix checking ep busy for CFC
  xhci: Workaround to get Intel xHCI reset working more reliably
  usb: chipidea: imx: fix a possible NULL dereference
  usb: chipidea: usbmisc_imx: fix a possible NULL dereference
  usb: chipidea: otg: gadget module load and unload support
  usb: chipidea: debug: disable usb irq while role switch
  ARM: dts: imx27.dtsi: change the clock information for usb
  usb: chipidea: imx: refine clock operations to adapt for all platforms
  usb: gadget: atmel_usba_udc: Expose correct device speed
  usb: musb: enable usb_dma parameter
  usb: phy: phy-mxs-usb: fix a possible NULL dereference
  usb: dwc3: gadget: let us set lower max_speed
  usb: musb: fix tx fifo flush handling
  usb: gadget: f_loopback: fix the warning during the enumeration
  usb: dwc2: host: Fix remote wakeup when not in DWC2_L2
  ...

9 years agoMerge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus
Linus Torvalds [Sun, 22 Nov 2015 20:59:46 +0000 (12:59 -0800)]
Merge branch 'upstream' of git://git.linux-mips.org/ralf/upstream-linus

Pull MIPS fixes from Ralf Baechle:

 - Fix a flood of annoying build warnings

 - A number of fixes for Atheros 79xx platforms

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MIPS: ath79: Add a machine entry for booting OF machines
  MIPS: ath79: Fix the size of the MISC INTC registers in ar9132.dtsi
  MIPS: ath79: Fix the DDR control initialization on ar71xx and ar934x
  MIPS: Fix flood of warnings about comparsion being always true.

9 years agoMerge branch 'parisc-4.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller...
Linus Torvalds [Sun, 22 Nov 2015 20:50:58 +0000 (12:50 -0800)]
Merge branch 'parisc-4.4-2' of git://git./linux/kernel/git/deller/parisc-linux

Pull parisc update from Helge Deller:
 "This patchset adds Huge Page and HUGETLBFS support for parisc"

Honestly, the hugepage support should have gone through in the merge
window, and is not really an rc-time fix.  But it only touches
arch/parisc, and I cannot find it in myself to care.  If one of the
three parisc users notices a breakage, I will point at Helge and make
rude farting noises.

* 'parisc-4.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Map kernel text and data on huge pages
  parisc: Add Huge Page and HUGETLBFS support
  parisc: Use long branch to do_syscall_trace_exit
  parisc: Increase initial kernel mapping to 32MB on 64bit kernel
  parisc: Initialize the fault vector earlier in the boot process.
  parisc: Add defines for Huge page support
  parisc: Drop unused MADV_xxxK_PAGES flags from asm/mman.h
  parisc: Drop definition of start_thread_som for HP-UX SOM binaries
  parisc: Fix wrong comment regarding first pmd entry flags

9 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 22 Nov 2015 20:37:20 +0000 (12:37 -0800)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf tool fixes from Thomas Gleixner:
 "A couple of fixes for perf tools:

   - Build system updates

   - Plug a memory leak in an error path of perf probe

   - Tear down probes correctly when adding fails

   - Fixes to the perf symbol handling

   - Fix ordering of event processing in buildid-list

   - Fix per DSO filtering in the histogram browser"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf probe: Clear probe_trace_event when add_probe_trace_event() fails
  perf probe: Fix memory leaking on failure by clearing all probe_trace_events
  perf inject: Also re-pipe lost_samples event
  perf buildid-list: Requires ordered events
  perf symbols: Fix dso lookup by long name and missing buildids
  perf symbols: Allow forcing reading of non-root owned files by root
  perf hists browser: The dso can be obtained from popup_action->ms.map->dso
  perf hists browser: Fix 'd' hotkey action to filter by DSO
  perf symbols: Rebuild rbtree when adjusting symbols for kcore
  tools: Add a "make all" rule
  tools: Actually install tmon in the install rule

9 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 22 Nov 2015 20:00:12 +0000 (12:00 -0800)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:
 "This update contains:

   - MPX updates for handling 32bit processes

   - A fix for a long standing bug in 32bit signal frame handling
     related to FPU/XSAVE state

   - Handle get_xsave_addr() correctly in KVM

   - Fix SMAP check under paravirtualization

   - Add a comment to the static function trace entry to avoid further
     confusion about the difference to dynamic tracing"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/cpu: Fix SMAP check in PVOPS environments
  x86/ftrace: Add comment on static function tracing
  x86/fpu: Fix get_xsave_addr() behavior under virtualization
  x86/fpu: Fix 32-bit signal frame handling
  x86/mpx: Fix 32-bit address space calculation
  x86/mpx: Do proper get_user() when running 32-bit binaries on 64-bit kernels

9 years agoslab/slub: adjust kmem_cache_alloc_bulk API
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:58 +0000 (15:57 -0800)]
slab/slub: adjust kmem_cache_alloc_bulk API

Adjust kmem_cache_alloc_bulk API before we have any real users.

Adjust API to return type 'int' instead of previously type 'bool'.  This
is done to allow future extension of the bulk alloc API.

A future extension could be to allow SLUB to stop at a page boundary, when
specified by a flag, and then return the number of objects.

The advantage of this approach, would make it easier to make bulk alloc
run without local IRQs disabled.  With an approach of cmpxchg "stealing"
the entire c->freelist or page->freelist.  To avoid overshooting we would
stop processing at a slab-page boundary.  Else we always end up returning
some objects at the cost of another cmpxchg.

To keep compatible with future users of this API linking against an older
kernel when using the new flag, we need to return the number of allocated
objects with this API change.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: add missing kmem cgroup support to kmem_cache_free_bulk
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:55 +0000 (15:57 -0800)]
slub: add missing kmem cgroup support to kmem_cache_free_bulk

Initial implementation missed support for kmem cgroup support in
kmem_cache_free_bulk() call, add this.

If CONFIG_MEMCG_KMEM is not enabled, the compiler should be smart enough
to not add any asm code.

Incoming bulk free objects can belong to different kmem cgroups, and
object free call can happen at a later point outside memcg context.  Thus,
we need to keep the orig kmem_cache, to correctly verify if a memcg object
match against its "root_cache" (s->memcg_params.root_cache).

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: fix kmem cgroup bug in kmem_cache_alloc_bulk
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:52 +0000 (15:57 -0800)]
slub: fix kmem cgroup bug in kmem_cache_alloc_bulk

The call slab_pre_alloc_hook() interacts with kmemgc and is not allowed to
be called several times inside the bulk alloc for loop, due to the call to
memcg_kmem_get_cache().

This would result in hitting the VM_BUG_ON in __memcg_kmem_get_cache.

As suggested by Vladimir Davydov, change slab_post_alloc_hook() to be able
to handle an array of objects.

A subtle detail is, loop iterator "i" in slab_post_alloc_hook() must have
same type (size_t) as size argument.  This helps the compiler to easier
realize that it can remove the loop, when all debug statements inside loop
evaluates to nothing.  Note, this is only an issue because the kernel is
compiled with GCC option: -fno-strict-overflow

In slab_alloc_node() the compiler inlines and optimizes the invocation of
slab_post_alloc_hook(s, flags, 1, &object) by removing the loop and access
object directly.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reported-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Suggested-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: optimize bulk slowpath free by detached freelist
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:49 +0000 (15:57 -0800)]
slub: optimize bulk slowpath free by detached freelist

This change focus on improving the speed of object freeing in the
"slowpath" of kmem_cache_free_bulk.

The calls slab_free (fastpath) and __slab_free (slowpath) have been
extended with support for bulk free, which amortize the overhead of
the (locked) cmpxchg_double.

To use the new bulking feature, we build what I call a detached
freelist.  The detached freelist takes advantage of three properties:

 1) the free function call owns the object that is about to be freed,
    thus writing into this memory is synchronization-free.

 2) many freelist's can co-exist side-by-side in the same slab-page
    each with a separate head pointer.

 3) it is the visibility of the head pointer that needs synchronization.

Given these properties, the brilliant part is that the detached
freelist can be constructed without any need for synchronization.  The
freelist is constructed directly in the page objects, without any
synchronization needed.  The detached freelist is allocated on the
stack of the function call kmem_cache_free_bulk.  Thus, the freelist
head pointer is not visible to other CPUs.

All objects in a SLUB freelist must belong to the same slab-page.
Thus, constructing the detached freelist is about matching objects
that belong to the same slab-page.  The bulk free array is scanned is
a progressive manor with a limited look-ahead facility.

Kmem debug support is handled in call of slab_free().

Notice kmem_cache_free_bulk no longer need to disable IRQs. This
only slowed down single free bulk with approx 3 cycles.

Performance data:
 Benchmarked[1] obj size 256 bytes on CPU i7-4790K @ 4.00GHz

SLUB fastpath single object quick reuse: 47 cycles(tsc) 11.931 ns

To get stable and comparable numbers, the kernel have been booted with
"slab_merge" (this also improve performance for larger bulk sizes).

Performance data, compared against fallback bulking:

bulk -  fallback bulk            - improvement with this patch
   1 -  62 cycles(tsc) 15.662 ns - 49 cycles(tsc) 12.407 ns- improved 21.0%
   2 -  55 cycles(tsc) 13.935 ns - 30 cycles(tsc) 7.506 ns - improved 45.5%
   3 -  53 cycles(tsc) 13.341 ns - 23 cycles(tsc) 5.865 ns - improved 56.6%
   4 -  52 cycles(tsc) 13.081 ns - 20 cycles(tsc) 5.048 ns - improved 61.5%
   8 -  50 cycles(tsc) 12.627 ns - 18 cycles(tsc) 4.659 ns - improved 64.0%
  16 -  49 cycles(tsc) 12.412 ns - 17 cycles(tsc) 4.495 ns - improved 65.3%
  30 -  49 cycles(tsc) 12.484 ns - 18 cycles(tsc) 4.533 ns - improved 63.3%
  32 -  50 cycles(tsc) 12.627 ns - 18 cycles(tsc) 4.707 ns - improved 64.0%
  34 -  96 cycles(tsc) 24.243 ns - 23 cycles(tsc) 5.976 ns - improved 76.0%
  48 -  83 cycles(tsc) 20.818 ns - 21 cycles(tsc) 5.329 ns - improved 74.7%
  64 -  74 cycles(tsc) 18.700 ns - 20 cycles(tsc) 5.127 ns - improved 73.0%
 128 -  90 cycles(tsc) 22.734 ns - 27 cycles(tsc) 6.833 ns - improved 70.0%
 158 -  99 cycles(tsc) 24.776 ns - 30 cycles(tsc) 7.583 ns - improved 69.7%
 250 - 104 cycles(tsc) 26.089 ns - 37 cycles(tsc) 9.280 ns - improved 64.4%

Performance data, compared current in-kernel bulking:

bulk - curr in-kernel  - improvement with this patch
   1 -  46 cycles(tsc) - 49 cycles(tsc) - improved (cycles:-3) -6.5%
   2 -  27 cycles(tsc) - 30 cycles(tsc) - improved (cycles:-3) -11.1%
   3 -  21 cycles(tsc) - 23 cycles(tsc) - improved (cycles:-2) -9.5%
   4 -  18 cycles(tsc) - 20 cycles(tsc) - improved (cycles:-2) -11.1%
   8 -  17 cycles(tsc) - 18 cycles(tsc) - improved (cycles:-1) -5.9%
  16 -  18 cycles(tsc) - 17 cycles(tsc) - improved (cycles: 1)  5.6%
  30 -  18 cycles(tsc) - 18 cycles(tsc) - improved (cycles: 0)  0.0%
  32 -  18 cycles(tsc) - 18 cycles(tsc) - improved (cycles: 0)  0.0%
  34 -  78 cycles(tsc) - 23 cycles(tsc) - improved (cycles:55) 70.5%
  48 -  60 cycles(tsc) - 21 cycles(tsc) - improved (cycles:39) 65.0%
  64 -  49 cycles(tsc) - 20 cycles(tsc) - improved (cycles:29) 59.2%
 128 -  69 cycles(tsc) - 27 cycles(tsc) - improved (cycles:42) 60.9%
 158 -  79 cycles(tsc) - 30 cycles(tsc) - improved (cycles:49) 62.0%
 250 -  86 cycles(tsc) - 37 cycles(tsc) - improved (cycles:49) 57.0%

Performance with normal SLUB merging is significantly slower for
larger bulking.  This is believed to (primarily) be an effect of not
having to share the per-CPU data-structures, as tuning per-CPU size
can achieve similar performance.

bulk - slab_nomerge   -  normal SLUB merge
   1 -  49 cycles(tsc) - 49 cycles(tsc) - merge slower with cycles:0
   2 -  30 cycles(tsc) - 30 cycles(tsc) - merge slower with cycles:0
   3 -  23 cycles(tsc) - 23 cycles(tsc) - merge slower with cycles:0
   4 -  20 cycles(tsc) - 20 cycles(tsc) - merge slower with cycles:0
   8 -  18 cycles(tsc) - 18 cycles(tsc) - merge slower with cycles:0
  16 -  17 cycles(tsc) - 17 cycles(tsc) - merge slower with cycles:0
  30 -  18 cycles(tsc) - 23 cycles(tsc) - merge slower with cycles:5
  32 -  18 cycles(tsc) - 22 cycles(tsc) - merge slower with cycles:4
  34 -  23 cycles(tsc) - 22 cycles(tsc) - merge slower with cycles:-1
  48 -  21 cycles(tsc) - 22 cycles(tsc) - merge slower with cycles:1
  64 -  20 cycles(tsc) - 48 cycles(tsc) - merge slower with cycles:28
 128 -  27 cycles(tsc) - 57 cycles(tsc) - merge slower with cycles:30
 158 -  30 cycles(tsc) - 59 cycles(tsc) - merge slower with cycles:29
 250 -  37 cycles(tsc) - 56 cycles(tsc) - merge slower with cycles:19

Joint work with Alexander Duyck.

[1] https://github.com/netoptimizer/prototype-kernel/blob/master/kernel/mm/slab_bulk_test01.c

[akpm@linux-foundation.org: BUG_ON -> WARN_ON;return]
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: support for bulk free with SLUB freelists
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:46 +0000 (15:57 -0800)]
slub: support for bulk free with SLUB freelists

Make it possible to free a freelist with several objects by adjusting API
of slab_free() and __slab_free() to have head, tail and an objects counter
(cnt).

Tail being NULL indicate single object free of head object.  This allow
compiler inline constant propagation in slab_free() and
slab_free_freelist_hook() to avoid adding any overhead in case of single
object free.

This allows a freelist with several objects (all within the same
slab-page) to be free'ed using a single locked cmpxchg_double in
__slab_free() and with an unlocked cmpxchg_double in slab_free().

Object debugging on the free path is also extended to handle these
freelists.  When CONFIG_SLUB_DEBUG is enabled it will also detect if
objects don't belong to the same slab-page.

These changes are needed for the next patch to bulk free the detached
freelists it introduces and constructs.

Micro benchmarking showed no performance reduction due to this change,
when debugging is turned off (compiled with CONFIG_SLUB_DEBUG).

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoparisc: Map kernel text and data on huge pages
Helge Deller [Sat, 21 Nov 2015 23:07:44 +0000 (00:07 +0100)]
parisc: Map kernel text and data on huge pages

Adjust the linker script and map_pages() to map kernel text and data on
physical 1MB huge/large pages.

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Add Huge Page and HUGETLBFS support
Helge Deller [Sat, 21 Nov 2015 23:07:06 +0000 (00:07 +0100)]
parisc: Add Huge Page and HUGETLBFS support

This patch adds huge page support to allow userspace to allocate huge
pages and to use hugetlbfs filesystem on 32- and 64-bit Linux kernels.
A later patch will add kernel support to map kernel text and data on
huge pages.

The only requirement is, that the kernel needs to be compiled for a
PA8X00 CPU (PA2.0 architecture). Older PA1.X CPUs do not support
variable page sizes. 64bit Kernels are compiled for PA2.0 by default.

Technically on parisc multiple physical huge pages may be needed to
emulate standard 2MB huge pages.

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Use long branch to do_syscall_trace_exit
Helge Deller [Fri, 20 Nov 2015 10:22:32 +0000 (11:22 +0100)]
parisc: Use long branch to do_syscall_trace_exit

Use the 22bit instead of the 17bit branch instruction on a 64bit kernel
to reach the do_syscall_trace_exit function from the gateway page.
A huge page enabled kernel may need the additional branch distance bits.

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Increase initial kernel mapping to 32MB on 64bit kernel
Helge Deller [Fri, 20 Nov 2015 10:17:27 +0000 (11:17 +0100)]
parisc: Increase initial kernel mapping to 32MB on 64bit kernel

For the 64bit kernel the initially 16 MB kernel memory might become too
small if you build a kernel with many modules built-in and with kernel
text and data areas mapped on huge pages.

This patch increases the initial mapping to 32MB for 64bit kernels and
keeps 16MB for 32bit kernels.

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Initialize the fault vector earlier in the boot process.
Helge Deller [Fri, 20 Nov 2015 09:50:01 +0000 (10:50 +0100)]
parisc: Initialize the fault vector earlier in the boot process.

A fault vector on parisc needs to be 2K aligned.  Furthermore the
checksum of the fault vector needs to sum up to 0 which is being
calculated and written at runtime.

Up to now we aligned both PA20 and PA11 fault vectors on the same 4K
page in order to easily write the checksum after having mapped the
kernel read-only (by mapping this page only as read-write).
But when we want to map the kernel text and data on huge pages this
makes things harder.
So, simplify it by aligning both fault vectors on 2K boundries and write
the checksum before we map the page read-only.

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Add defines for Huge page support
Helge Deller [Fri, 20 Nov 2015 14:46:52 +0000 (15:46 +0100)]
parisc: Add defines for Huge page support

Huge pages on parisc will have the same size as one pmd table, which
is on a 64bit kernel 2MB on a kernel with 4K kernel page sizes, and
on a 32bit kernel 4MB when used with 4K kernel pages.

Since parisc does not physically supports 2MB huge page sizes, emulate
it with two consecutive 1MB page sizes instead. Keeping the same huge
page size as one pmd will allow us to add transparent huge page support
later on.

Bit 21 in the pte flags was unused and will now be used to mark a page
as huge page (_PAGE_HPAGE_BIT).

Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoparisc: Drop unused MADV_xxxK_PAGES flags from asm/mman.h
Helge Deller [Sun, 22 Nov 2015 11:14:14 +0000 (12:14 +0100)]
parisc: Drop unused MADV_xxxK_PAGES flags from asm/mman.h

Drop the MADV_xxK_PAGES flags, which were never used and were from a proposed
API which was never integrated into the generic Linux kernel code.

Cc: stable@vger.kernel.org
Signed-off-by: Helge Deller <deller@gmx.de>
9 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Sat, 21 Nov 2015 18:49:13 +0000 (10:49 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "A bunch of fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  slub: mark the dangling ifdef #else of CONFIG_SLUB_DEBUG
  slub: avoid irqoff/on in bulk allocation
  slub: create new ___slab_alloc function that can be called with irqs disabled
  mm: fix up sparse warning in gfpflags_allow_blocking
  ocfs2: fix umask ignored issue
  PM/OPP: add entry in MAINTAINERS
  kernel/panic.c: turn off locks debug before releasing console lock
  kernel/signal.c: unexport sigsuspend()
  kasan: fix kmemleak false-positive in kasan_module_alloc()
  fat: fix fake_offset handling on error path
  mm/hugetlbfs: fix bugs in fallocate hole punch of areas with holes
  mm/page-writeback.c: initialize m_dirty to avoid compile warning
  various: fix pci_set_dma_mask return value checking
  mm: loosen MADV_NOHUGEPAGE to enable Qemu postcopy on s390
  mm: vmalloc: don't remove inexistent guard hole in remove_vm_area()
  tools/vm/page-types.c: support KPF_IDLE
  ncpfs: don't allow negative timeouts
  configfs: allow dynamic group creation
  MAINTAINERS: add Moritz as reviewer for FPGA Manager Framework
  slab.h: sprinkle __assume_aligned attributes

9 years agoMerge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 21 Nov 2015 18:26:24 +0000 (10:26 -0800)]
Merge branch 'timers-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull timer fixes from Thomas Gleixner:
 "Two timer fixlets from Arnd:

   - Use proper constant size in the FSL timer driver
   - Prevent a build error for legacy platforms"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  clocksource: Disallow drivers for ARCH_USES_GETTIMEOFFSET
  clocksource/fsl: Avoid harmless 64-bit warnings

9 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 21 Nov 2015 18:19:15 +0000 (10:19 -0800)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq fixes from Thomas Gleixner:
 "Three fixes for the ARM GIC interrupt controller from Marc addressing
  various shortcomings versus boot initialization and suspend/resume"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic: Add save/restore of the active state
  irqchip/gic: Clear enable bits before restoring them
  irqchip/gic: Make sure all interrupts are deactivated at boot

9 years agoMerge tag 'for-linus-20151120' of git://git.infradead.org/linux-mtd
Linus Torvalds [Sat, 21 Nov 2015 17:52:07 +0000 (09:52 -0800)]
Merge tag 'for-linus-20151120' of git://git.infradead.org/linux-mtd

Pull MTD fixes from Brian Norris:

 - MAINTAINERS updates for brcmnand driver

 - Fix reboot hangs seen when multiple NAND flash chips are registered
   with the same controller

 - Fix build issues on jz4740 NAND driver; the error was introduced in
   4.3, so I guess nobody really cared, but we might as well fix it

* tag 'for-linus-20151120' of git://git.infradead.org/linux-mtd:
  MAINTAINERS: brcmnand: Add co-maintainer for Broadcom SoCs
  MAINTAINERS: brcmnand: Add Broadcom internal mailing-list
  mtd: nand: fix shutdown/reboot for multi-chip systems
  mtd: jz4740_nand: fix build on jz4740 after removing gpio.h

9 years agoserial: export fsl8250_handle_irq
Arnd Bergmann [Mon, 16 Nov 2015 15:48:09 +0000 (16:48 +0100)]
serial: export fsl8250_handle_irq

fsl8250_handle_irq is now used by the of_serial driver, and that fails
if it is a loadable module:

ERROR: "fsl8250_handle_irq" [drivers/tty/serial/of_serial.ko] undefined!

This exports the symbol to avoid randconfig errors.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: d43b54d269d2 ("serial: Enable Freescale 16550 workaround on arm")
Cc: Scott Wood <scottwood@freescale.com>
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agoserial: 8250_mid: Add missing dependency
Heikki Krogerus [Thu, 12 Nov 2015 13:21:23 +0000 (15:21 +0200)]
serial: 8250_mid: Add missing dependency

8250_mid uses rational_best_approximation() function, so the
driver needs to select CONFIG_RATIONAL option.

This fixes build error when CONFIG_RATIONAL is not enabled:

drivers/built-in.o: In function `mid8250_set_termios':
8250_mid.c:(.text+0x10169a): undefined reference to `rational_best_approximation'

Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agotty: audit: Fix audit source
Peter Hurley [Sun, 8 Nov 2015 13:52:31 +0000 (08:52 -0500)]
tty: audit: Fix audit source

The data to audit/record is in the 'from' buffer (ie., the input
read buffer).

Fixes: 72586c6061ab ("n_tty: Fix auditing support for cannonical mode")
Cc: stable <stable@vger.kernel.org> # 4.1+
Cc: Miloslav Trmač <mitr@redhat.com>
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Laura Abbott <labbott@fedoraproject.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agoserial: etraxfs-uart: Fix crash
Guenter Roeck [Mon, 2 Nov 2015 02:32:56 +0000 (18:32 -0800)]
serial: etraxfs-uart: Fix crash

Since commit 7d8c70d8048c ("serial: mctrl-gpio: rename init function"),
crisv32 either do not build or crash as follows.

Unable to handle kernel NULL pointer dereference
Linux 4.3.0-rc7-next-20151101 #1 Sun Nov 1 11:41:28 PST 2015
...
Call Trace: [<c0004a0e>] show_stack+0x0/0x9e
[<c004c0c0>] printk+0x0/0x2c
[<c00059d4>] show_registers+0x14a/0x1c2
[<c004c0c0>] printk+0x0/0x2c
[<c0004b52>] die_if_kernel+0x7c/0x9e
[<c0005346>] do_page_fault+0x32e/0x3e6
[<c01dc59c>] of_get_property+0x0/0x2c
[<c01e0558>] of_irq_parse_raw+0x12a/0x376
[<c01dc59c>] of_get_property+0x0/0x2c
[<c0053aca>] get_page_from_freelist+0x73e/0x856
[<c01dc59c>] of_get_property+0x0/0x2c
[<c0008912>] d_mmu_refill+0x10a/0x112
[<c01b488c>] devm_kmalloc+0x40/0x56
[<c01b47d0>] add_dr+0xc/0x1c
[<c01b4800>] devm_add_action+0x2/0x4e
[<c01abdbc>] mctrl_gpio_init_noauto+0x1c/0x76
[<c01abf9e>] mctrl_gpio_init+0x22/0x110

The function call in the etraxfs-uart driver was not renamed,
possibly due to interference with commit 7b9c5162c182 ("serial:
etraxfs-uart: use mctrl_gpio helpers for handling modem signals").

Fixes: 7d8c70d8048c ("serial: mctrl-gpio: rename init function")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Niklas Cassel <nks@flawful.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agoserial: fsl_lpuart: Fix earlycon support
Peter Hurley [Tue, 20 Oct 2015 13:55:21 +0000 (09:55 -0400)]
serial: fsl_lpuart: Fix earlycon support

Earlycon support for Freescale lpuart should only be enabled when
console support is enabled.

Fixes: 1d59b382f1c4 ("serial: fsl_lpuart: add earlycon support")
Acked-by: Stefan Agner <stefan@agner.ch>
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agobcm63xx_uart: Use the device name when registering an interrupt
Simon Arlott [Sun, 15 Nov 2015 16:14:32 +0000 (16:14 +0000)]
bcm63xx_uart: Use the device name when registering an interrupt

Use the device name when registering an interrupt so that multiple
ports don't all have the same interrupt name.

Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agotty: Fix direct use of tty buffer work
Peter Hurley [Sun, 8 Nov 2015 12:53:06 +0000 (07:53 -0500)]
tty: Fix direct use of tty buffer work

Recent abstraction of tty buffer work introduced api to manage
tty input kworker; use it.

Fixes: e176058f0de5 ("tty: Abstract tty buffer work")
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agotty: Fix tty_send_xchar() lock order inversion
Peter Hurley [Wed, 11 Nov 2015 13:03:54 +0000 (08:03 -0500)]
tty: Fix tty_send_xchar() lock order inversion

The correct lock order is atomic_write_lock => termios_rwsem, as
established by tty_write() => n_tty_write().

Fixes: c274f6ef1c666 ("tty: Hold termios_rwsem for tcflow(TCIxxx)")
Reported-and-Tested-by: Dmitry Vyukov <dvyukov@google.com>
Cc: <stable@vger.kernel.org> # v3.18+
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agoslub: mark the dangling ifdef #else of CONFIG_SLUB_DEBUG
Jesper Dangaard Brouer [Fri, 20 Nov 2015 23:57:41 +0000 (15:57 -0800)]
slub: mark the dangling ifdef #else of CONFIG_SLUB_DEBUG

The #ifdef of CONFIG_SLUB_DEBUG is located very far from the associated
#else.  For readability mark it with a comment.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: avoid irqoff/on in bulk allocation
Christoph Lameter [Fri, 20 Nov 2015 23:57:38 +0000 (15:57 -0800)]
slub: avoid irqoff/on in bulk allocation

Use the new function that can do allocation while interrupts are disabled.
Avoids irq on/off sequences.

Signed-off-by: Christoph Lameter <cl@linux.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: create new ___slab_alloc function that can be called with irqs disabled
Christoph Lameter [Fri, 20 Nov 2015 23:57:35 +0000 (15:57 -0800)]
slub: create new ___slab_alloc function that can be called with irqs disabled

Bulk alloc needs a function like that because it enables interrupts before
calling __slab_alloc which promptly disables them again using the expensive
local_irq_save().

Signed-off-by: Christoph Lameter <cl@linux.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm: fix up sparse warning in gfpflags_allow_blocking
Jeff Layton [Fri, 20 Nov 2015 23:57:32 +0000 (15:57 -0800)]
mm: fix up sparse warning in gfpflags_allow_blocking

sparse says:

    include/linux/gfp.h:274:26: warning: incorrect type in return expression (different base types)
    include/linux/gfp.h:274:26:    expected bool
    include/linux/gfp.h:274:26:    got restricted gfp_t

...add a forced cast to silence the warning.

Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoocfs2: fix umask ignored issue
Junxiao Bi [Fri, 20 Nov 2015 23:57:30 +0000 (15:57 -0800)]
ocfs2: fix umask ignored issue

New created file's mode is not masked with umask, and this makes umask not
work for ocfs2 volume.

Fixes: 702e5bc ("ocfs2: use generic posix ACL infrastructure")
Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Gang He <ghe@suse.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoPM/OPP: add entry in MAINTAINERS
Viresh Kumar [Fri, 20 Nov 2015 23:57:27 +0000 (15:57 -0800)]
PM/OPP: add entry in MAINTAINERS

Add entry for operating performance points into MAINTAINERS file.  This
will also allow get_maintainers to list OPP stakeholders properly.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Stephen Boyd <sboyd@codeaurora.org>
Cc: Rafael Wysocki <rjw@rjwysocki.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agokernel/panic.c: turn off locks debug before releasing console lock
Vitaly Kuznetsov [Fri, 20 Nov 2015 23:57:24 +0000 (15:57 -0800)]
kernel/panic.c: turn off locks debug before releasing console lock

Commit 08d78658f393 ("panic: release stale console lock to always get the
logbuf printed out") introduced an unwanted bad unlock balance report when
panic() is called directly and not from OOPS (e.g.  from out_of_memory()).
The difference is that in case of OOPS we disable locks debug in
oops_enter() and on direct panic call nobody does that.

Fixes: 08d78658f393 ("panic: release stale console lock to always get the logbuf printed out")
Reported-by: kernel test robot <ying.huang@linux.intel.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Jiri Kosina <jkosina@suse.cz>
Cc: Baoquan He <bhe@redhat.com>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Xie XiuQi <xiexiuqi@huawei.com>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Petr Mladek <pmladek@suse.cz>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agokernel/signal.c: unexport sigsuspend()
Richard Weinberger [Fri, 20 Nov 2015 23:57:21 +0000 (15:57 -0800)]
kernel/signal.c: unexport sigsuspend()

sigsuspend() is nowhere used except in signal.c itself, so we can mark it
static do not pollute the global namespace.

But this patch is more than a boring cleanup patch, it fixes a real issue
on UserModeLinux.  UML has a special console driver to display ttys using
xterm, or other terminal emulators, on the host side.  Vegard reported
that sometimes UML is unable to spawn a xterm and he's facing the
following warning:

  WARNING: CPU: 0 PID: 908 at include/linux/thread_info.h:128 sigsuspend+0xab/0xc0()

It turned out that this warning makes absolutely no sense as the UML
xterm code calls sigsuspend() on the host side, at least it tries.  But
as the kernel itself offers a sigsuspend() symbol the linker choose this
one instead of the glibc wrapper.  Interestingly this code used to work
since ever but always blocked signals on the wrong side.  Some recent
kernel change made the WARN_ON() trigger and uncovered the bug.

It is a wonderful example of how much works by chance on computers. :-)

Fixes: 68f3f16d9ad0f1 ("new helper: sigsuspend()")
Signed-off-by: Richard Weinberger <richard@nod.at>
Reported-by: Vegard Nossum <vegard.nossum@oracle.com>
Tested-by: Vegard Nossum <vegard.nossum@oracle.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@vger.kernel.org> [3.5+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agokasan: fix kmemleak false-positive in kasan_module_alloc()
Andrey Ryabinin [Fri, 20 Nov 2015 23:57:18 +0000 (15:57 -0800)]
kasan: fix kmemleak false-positive in kasan_module_alloc()

Kmemleak reports the following leak:

unreferenced object 0xfffffbfff41ea000 (size 20480):
comm "modprobe", pid 65199, jiffies 4298875551 (age 542.568s)
hex dump (first 32 bytes):
  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
backtrace:
  [<ffffffff82354f5e>] kmemleak_alloc+0x4e/0xc0
  [<ffffffff8152e718>] __vmalloc_node_range+0x4b8/0x740
  [<ffffffff81574072>] kasan_module_alloc+0x72/0xc0
  [<ffffffff810efe68>] module_alloc+0x78/0xb0
  [<ffffffff812f6a24>] module_alloc_update_bounds+0x14/0x70
  [<ffffffff812f8184>] layout_and_allocate+0x16f4/0x3c90
  [<ffffffff812faa1f>] load_module+0x2ff/0x6690
  [<ffffffff813010b6>] SyS_finit_module+0x136/0x170
  [<ffffffff8239bbc9>] system_call_fastpath+0x16/0x1b
  [<ffffffffffffffff>] 0xffffffffffffffff

kasan_module_alloc() allocates shadow memory for module and frees it on
module unloading.  It doesn't store the pointer to allocated shadow memory
because it could be calculated from the shadowed address, i.e.
kasan_mem_to_shadow(addr).

Since kmemleak cannot find pointer to allocated shadow, it thinks that
memory leaked.

Use kmemleak_ignore() to tell kmemleak that this is not a leak and shadow
memory doesn't contain any pointers.

Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agofat: fix fake_offset handling on error path
OGAWA Hirofumi [Fri, 20 Nov 2015 23:57:15 +0000 (15:57 -0800)]
fat: fix fake_offset handling on error path

For the root directory, .  and ..  are faked (using dir_emit_dots()) and
ctx->pos is reset from 2 to 0.

A corrupted root directory could cause fat_get_entry() to fail, but
->iterate() (fat_readdir()) reports progress to the VFS (with ctx->pos
rewound to 0), so any following calls to ->iterate() continue to return
the same entries again and again.

The result is that userspace will never see the end of the directory,
causing e.g.  'ls' to hang in a getdents() loop.

[hirofumi@mail.parknet.co.jp: cleanup and make sure to correct fake_offset]
Reported-by: Vegard Nossum <vegard.nossum@oracle.com>
Tested-by: Vegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: Richard Weinberger <richard.weinberger@gmail.com>
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm/hugetlbfs: fix bugs in fallocate hole punch of areas with holes
Mike Kravetz [Fri, 20 Nov 2015 23:57:13 +0000 (15:57 -0800)]
mm/hugetlbfs: fix bugs in fallocate hole punch of areas with holes

Hugh Dickins pointed out problems with the new hugetlbfs fallocate hole
punch code.  These problems are in the routine remove_inode_hugepages and
mostly occur in the case where there are holes in the range of pages to be
removed.  These holes could be the result of a previous hole punch or
simply sparse allocation.  The current code could access pages outside the
specified range.

remove_inode_hugepages handles both hole punch and truncate operations.
Page index handling was fixed/cleaned up so that the loop index always
matches the page being processed.  The code now only makes a single pass
through the range of pages as it was determined page faults could not race
with truncate.  A cond_resched() was added after removing up to
PAGEVEC_SIZE pages.

Some totally unnecessary code in hugetlbfs_fallocate() that remained from
early development was also removed.

Tested with fallocate tests submitted here:
http://librelist.com/browser//libhugetlbfs/2015/6/25/patch-tests-add-tests-for-fallocate-system-call/
And, some ftruncate tests under development

Fixes: b5cec28d36f5 ("hugetlbfs: truncate_hugepages() takes a range of pages")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: "Hillf Danton" <hillf.zj@alibaba-inc.com>
Cc: <stable@vger.kernel.org> [4.3]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm/page-writeback.c: initialize m_dirty to avoid compile warning
Yang Shi [Fri, 20 Nov 2015 23:57:10 +0000 (15:57 -0800)]
mm/page-writeback.c: initialize m_dirty to avoid compile warning

When building kernel with gcc 5.2, the below warning is raised:

  mm/page-writeback.c: In function 'balance_dirty_pages.isra.10':
  mm/page-writeback.c:1545:17: warning: 'm_dirty' may be used uninitialized in this function [-Wmaybe-uninitialized]
     unsigned long m_dirty, m_thresh, m_bg_thresh;

The m_dirty{thresh, bg_thresh} are initialized in the block of "if
(mdtc)", so if mdts is null, they won't be initialized before being used.
Initialize m_dirty to zero, also initialize m_thresh and m_bg_thresh to
keep consistency.

They are used later by if condition: !mdtc || m_dirty <=
dirty_freerun_ceiling(m_thresh, m_bg_thresh)

If mdtc is null, dirty_freerun_ceiling will not be called at all, so the
initialization will not change any behavior other than just ceasing the
compile warning.

(akpm: the patch actually reduces .text size by ~20 bytes on gcc-4.x.y)

[akpm@linux-foundation.org: add comment]
Signed-off-by: Yang Shi <yang.shi@linaro.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agovarious: fix pci_set_dma_mask return value checking
Christoph Hellwig [Fri, 20 Nov 2015 23:57:07 +0000 (15:57 -0800)]
various: fix pci_set_dma_mask return value checking

pci_set_dma_mask returns a negative errno value, not a bool like
pci_dma_supported.  This of course was just a giant test for attention :)

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reported-by: Jongman Heo <jongman.heo@samsung.com>
Tested-by: Jongman Heo <jongman.heo@samsung.com> [pcnet32]
Acked-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Cc: Hans Verkuil <hverkuil@xs4all.nl>
Cc: Antti Palosaari <crope@iki.fi>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm: loosen MADV_NOHUGEPAGE to enable Qemu postcopy on s390
Jason J. Herne [Fri, 20 Nov 2015 23:57:04 +0000 (15:57 -0800)]
mm: loosen MADV_NOHUGEPAGE to enable Qemu postcopy on s390

MADV_NOHUGEPAGE processing is too restrictive.  kvm already disables
hugepage but hugepage_madvise() takes the error path when we ask to turn
on the MADV_NOHUGEPAGE bit and the bit is already on.  This causes Qemu's
new postcopy migration feature to fail on s390 because its first action is
to madvise the guest address space as NOHUGEPAGE.  This patch modifies the
code so that the operation succeeds without error now.

For consistency reasons do the same for MADV_HUGEPAGE.

Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com>
Reviewed-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm: vmalloc: don't remove inexistent guard hole in remove_vm_area()
Jerome Marchand [Fri, 20 Nov 2015 23:57:02 +0000 (15:57 -0800)]
mm: vmalloc: don't remove inexistent guard hole in remove_vm_area()

Commit 71394fe50146 ("mm: vmalloc: add flag preventing guard hole
allocation") missed a spot.  Currently remove_vm_area() decreases vm->size
to "remove" the guard hole page, even when it isn't present.  All but one
users just free the vm_struct rigth away and never access vm->size anyway.

Don't touch the size in remove_vm_area() and have __vunmap() use the
proper get_vm_area_size() helper.

Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Acked-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agotools/vm/page-types.c: support KPF_IDLE
Naoya Horiguchi [Fri, 20 Nov 2015 23:56:59 +0000 (15:56 -0800)]
tools/vm/page-types.c: support KPF_IDLE

PageIdle is exported in include/uapi/linux/kernel-page-flags.h, so let's
make page-types.c tool handle it.

Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoncpfs: don't allow negative timeouts
Dan Carpenter [Fri, 20 Nov 2015 23:56:56 +0000 (15:56 -0800)]
ncpfs: don't allow negative timeouts

This code causes a static checker warning because it's a user controlled
variable where we cap the upper bound but not the lower bound.  Let's
return an -EINVAL for negative timeouts.

[akpm@linux-foundation.org: remove unneeded `else']
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Jan Kara <jack@suse.com>
Cc: Petr Vandrovec <petr@vandrovec.name>
Cc: David Howells <dhowells@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoconfigfs: allow dynamic group creation
Daniel Baluta [Fri, 20 Nov 2015 23:56:53 +0000 (15:56 -0800)]
configfs: allow dynamic group creation

This patchset introduces IIO software triggers, offers a way of configuring
them via configfs and adds the IIO hrtimer based interrupt source to be used
with software triggers.

The architecture is now split in 3 parts, to remove all IIO trigger specific
parts from IIO configfs core:

(1) IIO configfs - creates the root of the IIO configfs subsys.
(2) IIO software triggers - software trigger implementation, dynamically
    creating /config/iio/triggers group.
(3) IIO hrtimer trigger - is the first interrupt source for software triggers
    (with syfs to follow). Each trigger type can implement its own set of
    attributes.

Lockdep seems to be happy with the locking in configfs patch.

This patch (of 5):

We don't want to hardcode default groups at subsystem
creation time. We export:
* configfs_register_group
* configfs_unregister_group
to allow drivers to programatically create/destroy groups
later, after module init time.

This is needed for IIO configfs support.

(akpm: the other 4 patches to be merged via the IIO tree)

Signed-off-by: Daniel Baluta <daniel.baluta@intel.com>
Suggested-by: Lars-Peter Clausen <lars@metafoo.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Joel Becker <jlbec@evilplan.org>
Cc: Hartmut Knaack <knaack.h@gmx.de>
Cc: Octavian Purdila <octavian.purdila@intel.com>
Cc: Paul Bolle <pebolle@tiscali.nl>
Cc: Adriana Reus <adriana.reus@intel.com>
Cc: Cristina Opriceana <cristina.opriceana@gmail.com>
Cc: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMAINTAINERS: add Moritz as reviewer for FPGA Manager Framework
Moritz Fischer [Fri, 20 Nov 2015 23:56:51 +0000 (15:56 -0800)]
MAINTAINERS: add Moritz as reviewer for FPGA Manager Framework

Nominate myself as Reviewer.

Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com>
Acked-by: Alan Tull <atull@opensource.altera.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslab.h: sprinkle __assume_aligned attributes
Rasmus Villemoes [Fri, 20 Nov 2015 23:56:48 +0000 (15:56 -0800)]
slab.h: sprinkle __assume_aligned attributes

The various allocators return aligned memory.  Telling the compiler that
allows it to generate better code in many cases, for example when the
return value is immediately passed to memset().

Some code does become larger, but at least we win twice as much as we lose:

$ scripts/bloat-o-meter /tmp/vmlinux vmlinux
add/remove: 0/0 grow/shrink: 13/52 up/down: 995/-2140 (-1145)

An example of the different (and smaller) code can be seen in mm_alloc(). Before:

:       48 8d 78 08             lea    0x8(%rax),%rdi
:       48 89 c1                mov    %rax,%rcx
:       48 89 c2                mov    %rax,%rdx
:       48 c7 00 00 00 00 00    movq   $0x0,(%rax)
:       48 c7 80 48 03 00 00    movq   $0x0,0x348(%rax)
:       00 00 00 00
:       31 c0                   xor    %eax,%eax
:       48 83 e7 f8             and    $0xfffffffffffffff8,%rdi
:       48 29 f9                sub    %rdi,%rcx
:       81 c1 50 03 00 00       add    $0x350,%ecx
:       c1 e9 03                shr    $0x3,%ecx
:       f3 48 ab                rep stos %rax,%es:(%rdi)

After:

:       48 89 c2                mov    %rax,%rdx
:       b9 6a 00 00 00          mov    $0x6a,%ecx
:       31 c0                   xor    %eax,%eax
:       48 89 d7                mov    %rdx,%rdi
:       f3 48 ab                rep stos %rax,%es:(%rdi)

So gcc's strategy is to do two possibly (but not really, of course)
unaligned stores to the first and last word, then do an aligned rep stos
covering the middle part with a little overlap.  Maybe arches which do not
allow unaligned stores gain even more.

I don't know if gcc can actually make use of alignments greater than 8 for
anything, so one could probably drop the __assume_xyz_alignment macros and
just use __assume_aligned(8).

The increases in code size are mostly caused by gcc deciding to
opencode strlen() using the check-four-bytes-at-a-time trick when it
knows the buffer is sufficiently aligned (one function grew by 200
bytes). Now it turns out that many of these strlen() calls showing up
were in fact redundant, and they're gone from -next. Applying the two
patches to next-20151001 bloat-o-meter instead says

add/remove: 0/0 grow/shrink: 6/52 up/down: 244/-2140 (-1896)

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMerge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa...
Linus Torvalds [Fri, 20 Nov 2015 23:52:52 +0000 (15:52 -0800)]
Merge branch 'i2c/for-current' of git://git./linux/kernel/git/wsa/linux

Pull i2c fixes from Wolfram Sang:
 "A few bugfixes and one PCI ID addition from I2C"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: i801: add Intel Lewisburg device IDs
  i2c: fix wakeup irq parsing
  i2c: xiic: Prevent concurrent running of the IRQ handler and __xiic_start_xfer()
  i2c: Revert "i2c: xiic: Do not reset controller before every transfer"
  i2c: imx: fix a compiling error

9 years agoMerge tag '4.4-fix' of git://git.lwn.net/linux
Linus Torvalds [Fri, 20 Nov 2015 23:47:27 +0000 (15:47 -0800)]
Merge tag '4.4-fix' of git://git.lwn.net/linux

Pull documentation fix from Jon Corbet:
 "A single fix from Mauro for a 4.4 regression that would cause the docs
  build to fail on systems with ancient Perl installations"

* tag '4.4-fix' of git://git.lwn.net/linux:
  kernel-doc: Make it compatible with Perl versions below 5.12 again

9 years agoMerge branch 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdim...
Linus Torvalds [Fri, 20 Nov 2015 23:00:50 +0000 (15:00 -0800)]
Merge branch 'libnvdimm-fixes' of git://git./linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:

 - A collection of crash and deadlock fixes for DAX that are also tagged
   for -stable.  We will look to re-enable DAX pmd mappings in 4.5, but
   for now 4.4 and -stable should disable it by default.

 - A fixup to ext2 and ext4 to mirror the same warning emitted by XFS
   when mounting with "-o dax"

* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  block: protect rw_page against device teardown
  mm, dax: fix DAX deadlocks (COW fault)
  dax: disable pmd mappings
  ext2, ext4: warn when mounting with dax enabled

9 years agoMerge tag 'pm+acpi-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 20 Nov 2015 17:01:07 +0000 (09:01 -0800)]
Merge tag 'pm+acpi-4.4-rc2' of git://git./linux/kernel/git/rafael/linux-pm

Pull more power management and ACPI updates from Rafael Wysocki:
 "These are mostly fixes and cleanups (ACPI core, PM core, cpufreq, ACPI
  EC driver, device properties) including three reverts of recent
  intel_pstate driver commits due to a regression introduced by one of
  them plus support for Atom Airmont cores in intel_pstate (which really
  boils down to adding new frequency tables for Airmont) and additional
  turbostat updates.

  Specifics:

   - Revert three recent intel_pstate driver commits one of which
     introduced a regression and the remaining two depend on the
     problematic one (Rafael Wysocki).

   - Fix breakage related to the recently introduced ACPI _CCA object
     support in the PCI DMA setup code (Suravee Suthikulpanit).

   - Fix up the recently introduced ACPI CPPC support to only use the
     hardware-reduced version of the PCCT structure as the only
     architecture to support it (ARM64) will only use hardware-reduced
     ACPI anyway (Ashwin Chaugule).

   - Fix a cpufreq mediatek driver build problem (Arnd Bergmann).

   - Fix the SMBus transaction handling implementation in the ACPI core
     to avoid re-entrant calls to wait_event_timeout() which makes
     intermittent boot stalls related to the Smart Battery Subsystem
     initialization go away and revert a workaround of another problem
     with the same underlying root cause (Chris Bainbridge).

   - Fix the generic wakeup interrupts framework to avoid using invalid
     IRQ numbers (Dmitry Torokhov).

   - Remove a redundant check from the ACPI EC driver (Markus Elfring).

   - Modify the intel_pstate driver so it can support more Atom flavors
     than just one (Baytrail) and add support for Atom Airmont cores
     (which require new freqnency tables) to it (Philippe Longepe).

   - Clean up MSR-related symbols in turbostat (Len Brown)"

* tag 'pm+acpi-4.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  PCI: Fix OF logic in pci_dma_configure()
  Revert "Documentation: kernel_parameters for Intel P state driver"
  cpufreq: mediatek: fix build error
  cpufreq: intel_pstate: Add separate support for Airmont cores
  cpufreq: intel_pstate: Replace BYT with ATOM
  Revert "cpufreq: intel_pstate: Use ACPI perf configuration"
  Revert "cpufreq: intel_pstate: Avoid calculation for max/min"
  ACPI-EC: Drop unnecessary check made before calling acpi_ec_delete_query()
  Revert "ACPI / SBS: Add 5 us delay to fix SBS hangs on MacBook"
  ACPI / SMBus: Fix boot stalls / high CPU caused by reentrant code
  PM / wakeirq: check that wake IRQ is valid before accepting it
  ACPI / CPPC: Use h/w reduced version of the PCCT structure
  x86: remove unused definition of MSR_NHM_PLATFORM_INFO
  tools/power turbostat: use new name for MSR_PLATFORM_INFO

9 years agoMerge tag 'powerpc-4.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Fri, 20 Nov 2015 16:20:04 +0000 (08:20 -0800)]
Merge tag 'powerpc-4.4-2' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixlet from Michael Ellerman:
 "Wire up sys_mlock2()"

* tag 'powerpc-4.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc: Wire up sys_mlock2()

9 years agoi2c: i801: add Intel Lewisburg device IDs
Alexandra Yates [Thu, 5 Nov 2015 19:40:25 +0000 (11:40 -0800)]
i2c: i801: add Intel Lewisburg device IDs

Adding Intel codename Lewisburg platform device IDs for SMBus.

Signed-off-by: Alexandra Yates <alexandra.yates@linux.intel.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
9 years agoi2c: fix wakeup irq parsing
Grygorii Strashko [Thu, 12 Nov 2015 13:42:26 +0000 (15:42 +0200)]
i2c: fix wakeup irq parsing

This patch fixes obvious copy-past error in wake up irq parsing
code which leads to the fact that dev_pm_set_wake_irq() will
be called with wrong IRQ number when "wakeup" IRQ is not
defined in DT.

Fixes: 3fffd1283927 ("i2c: allow specifying separate wakeup interrupt in device tree")
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Cc: <stable@vger.kernel.org> # v4.3
9 years agoi2c: xiic: Prevent concurrent running of the IRQ handler and __xiic_start_xfer()
Lars-Peter Clausen [Mon, 16 Nov 2015 13:42:04 +0000 (14:42 +0100)]
i2c: xiic: Prevent concurrent running of the IRQ handler and __xiic_start_xfer()

Prior to commit e6c9a037bc8a ("i2c: xiic: Remove the disabling of
interrupts") IRQs where disabled when the initial __xiic_start_xfer() was
called. After the commit the interrupt is enabled while the function is
running, this means it is possible for the interrupt to be triggered while
the function is still running. When this happens the internal data
structures get corrupted and undefined behavior can occur like the
following crash:

Internal error: Oops: 17 [#1] PREEMPT SMP ARM
Modules linked in:
CPU: 0 PID: 2040 Comm: i2cdetect Not tainted 4.0.0-02856-g047a308 #10956
Hardware name: Xilinx Zynq Platform
task: ee0c9500 ti: e99a2000 task.ti: e99a2000
PC is at __xiic_start_xfer+0x6c4/0x7c8
LR is at __xiic_start_xfer+0x690/0x7c8
pc : [<c02bbffc>]    lr : [<c02bbfc8>]    psr: 800f0013
sp : e99a3da8  ip : 00000000  fp : 00000000
r10: 00000001  r9 : 600f0013  r8 : f0180000
r7 : f0180000  r6 : c064e444  r5 : 00000017  r4 : ee031010
r3 : 00000000  r2 : 00000000  r1 : 600f0013  r0 : 0000000f
Flags: Nzcv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
Control: 18c5387d  Table: 29a5404a  DAC: 00000015
Process i2cdetect (pid: 2040, stack limit = 0xe99a2210)
Stack: (0xe99a3da8 to 0xe99a4000)
3da0:                   ee031010 00000000 00000001 ee031020 ee031224 c02bc5ec
3dc0: ee34c604 00000000 ee0c9500 e99a3dcc e99a3dd0 e99a3dd0 e99a3dd8 c069f0e8
3de0: 00000000 ee031020 c064e100 ffff90bb e99a3e48 c02b6590 ee031020 00000001
3e00: e99a3e48 ee031020 00000000 e99a3e63 00000001 c02b6ec4 00000000 00000000
3e20: 00000000 c02b7320 e99a3ef0 00000000 00000000 e99e3df0 00000000 00000000
3e40: 00000103 2814575f 0000003e c00a0000 e99a3e85 0001003e ee0c0000 e99a3e63
3e60: eefd3578 c064e61c ee0c9500 c0041e04 0000056c e9a56db8 00006e5a b6f5c000
3e80: ee0c9548 eefd0040 00000001 eefd3540 ee0c9500 eefd39a0 c064b540 ee0c9500
3ea0: 00000000 ee92b000 00000000 bef4862c ee34c600 e99ecdc0 00000720 00000003
3ec0: e99a2000 00000000 00000000 c02b8b30 00000000 00000000 00000000 e99a3f24
3ee0: b6e80000 00000000 00000000 c04257e8 00000000 e99a3f24 c02b8f08 00000703
3f00: 00000003 c02116bc ee935300 00000000 bef4862c ee34c600 e99ecdc0 c02b91f0
3f20: e99ecdc0 00000720 bef4862c eeb725f8 e99ecdc0 c00c9e2c 00000003 00000003
3f40: ee248dc0 00000000 ee248dc8 00000002 eeb7c1a8 00000000 00000000 c00bb360
3f60: 00000000 00000000 00000003 ee248dc0 bef4862c e99ecdc0 e99ecdc0 00000720
3f80: 00000003 e99a2000 00000000 c00c9f68 00000000 00000000 b6f22000 00000036
3fa0: c000dfa4 c000de20 00000000 00000000 00000003 00000720 bef4862c bef4862c
3fc0: 00000000 00000000 b6f22000 00000036 00000000 00000000 b6f60000 00000000
3fe0: 00013040 bef48614 00008cab b6ecdbe6 400f0030 00000003 2f7fd821 2f7fdc21
[<c02bbffc>] (__xiic_start_xfer) from [<c02bc5ec>] (xiic_xfer+0x94/0x168)
[<c02bc5ec>] (xiic_xfer) from [<c02b6590>] (__i2c_transfer+0x4c/0x7c)
[<c02b6590>] (__i2c_transfer) from [<c02b6ec4>] (i2c_transfer+0x9c/0xc4)
[<c02b6ec4>] (i2c_transfer) from [<c02b7320>] (i2c_smbus_xfer+0x3a0/0x4ec)
[<c02b7320>] (i2c_smbus_xfer) from [<c02b8b30>] (i2cdev_ioctl_smbus+0xb0/0x214)
[<c02b8b30>] (i2cdev_ioctl_smbus) from [<c02b91f0>] (i2cdev_ioctl+0xa0/0x1d4)
[<c02b91f0>] (i2cdev_ioctl) from [<c00c9e2c>] (do_vfs_ioctl+0x4b0/0x5b8)
[<c00c9e2c>] (do_vfs_ioctl) from [<c00c9f68>] (SyS_ioctl+0x34/0x5c)
[<c00c9f68>] (SyS_ioctl) from [<c000de20>] (ret_fast_syscall+0x0/0x34)
Code: e283300c e5843210 eafffe64 e5943210 (e1d320b4)

The issue can easily be reproduced by performing I2C access under high
system load or IO load.

To fix the issue protect the invocation to __xiic_start_xfer() form
xiic_start_xfer() with the same lock that is used to protect the interrupt
handler.

Fixes: e6c9a037bc8a ("i2c: xiic: Remove the disabling of interrupts")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Reviewed-by: Shubhrajyoti Datta <shubhraj@xilinx.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
9 years agoi2c: Revert "i2c: xiic: Do not reset controller before every transfer"
Lars-Peter Clausen [Mon, 16 Nov 2015 13:42:03 +0000 (14:42 +0100)]
i2c: Revert "i2c: xiic: Do not reset controller before every transfer"

Commit d701667bb331 ("i2c: xiic: Do not reset controller before every
transfer") removed the reinitialization of the controller before the start
of each transfer. Apparently this change is not safe to make and the commit
results in random I2C bus failures.

An easy way to trigger the issue is to run i2cdetect.

Without the patch applied:
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
 00:          -- -- -- -- -- -- -- -- -- -- -- -- --
 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 30: -- -- -- -- -- -- -- -- UU UU -- UU 3c -- -- UU
 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 70: -- -- -- -- -- -- -- --

With the patch applied every other or so invocation:
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
 00:          03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f
 10: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f
 20: 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f
 30: -- -- -- -- -- -- -- -- UU UU -- UU 3c -- -- UU
 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
 70: -- -- -- -- -- -- -- --

So revert the commit for now.

Fixes: d701667bb331 ("i2c: xiic: Do not reset controller before every transfer")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Shubhrajyoti Datta <shubhraj@xilinx.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
9 years agoi2c: imx: fix a compiling error
Hou Zhiqiang [Tue, 17 Nov 2015 09:53:18 +0000 (17:53 +0800)]
i2c: imx: fix a compiling error

drivers/i2c/busses/i2c-imx.c:978:2: error: implicit declaration of
function ‘pinctrl_select_state’ [-Werror=implicit-function-declaration]
  pinctrl_select_state(i2c_imx->pinctrl, i2c_imx->pinctrl_pins_gpio);
  ^

Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@freescale.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
9 years agoMIPS: ath79: Add a machine entry for booting OF machines
Alban Bedel [Tue, 17 Nov 2015 20:52:00 +0000 (21:52 +0100)]
MIPS: ath79: Add a machine entry for booting OF machines

As I'm using a board with a broken old bootloader I hardcoded the
mips_machtype and did't notice that the machine entry was still
missing.

[ralf@linux-mips.org: Fixed spelling message noticed by Sergei Shtylyov
<sergei.shtylyov@cogentembedded.com>.]

Signed-off-by: Alban Bedel <albeu@free.fr>
Cc: Qais Yousef <qais.yousef@imgtec.com>
Cc: Felix Fietkau <nbd@openwrt.org>
Cc: Andrew Bresticker <abrestic@chromium.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/11503/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>