When rounding a relative timeout we need to use round_jiffies_relative().
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Set bits 0, 4, 5 and 7 of PCI configuration register 0x40 in the
quirk.  This has the following effects and is recommended by the
vendor.

* Force enable of IDE channels (used to be left alone as BIOS
  configured)

* Change initial phase behavior of PIO cycle such that the host pulls
  down the bus instead of tristating it.  Vendor recommends this
  setting.

The above settings are better for the current generation of
controllers and needed for the upcoming next generation.

Tested on JMB363.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Ethan Hsiao <ethanhsiao@jmicron.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Another one doing spurious NCQ completions.  Blacklist it.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Luca Tettamanti <kronos.it@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Did a complete audit of these and found we have another error case.

ata_bus_softreset calls ata_check_status which means that it tries to do
an ioread8 on the port blindly and check versus 0xFF for an error.

It should of course be using the ap->ops method for this via chk_status,
and this bug causes a wrog status call on the NS87415 at least.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Tackle the relatively sane complaints of checkpatch --file.

The vast majority is indentation and whitespace changes, the rest are

* #include fixes
* printk KERN_xxx prefix addition
* BSS/initializer cleanups
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

	Update ALB mode monitor to hold correct locks (RTNL and nothing
else) when calling dev_set_promiscuity.
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

	Convert more lock acquisitions to _bh flavor to avoid deadlock
with workqueue activity and add acquisition of RTNL in appropriate places.
Affects ALB mode, as well as core bonding functions and sysfs.
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

	Convert locking-related activity to new & improved system.
Convert some lock acquisitions to _bh and rework parts of ALB mode, both
to avoid deadlocks with workqueue activity.
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

	Convert mii (link state) monitor to acquire correct locks for
failover events.  In particular, failovers generally require RTNL at a low
level (when manipulating device MAC addresses, for example) and no other
locks.  The high level monitor is responsible for acquiring a known set
of locks, RTNL, the bond->lock for read and the slave_lock for write, and
the low level failover processing can then release appropriate locks as
needed.  This patch provides the high level portion.

	As it is undesirable to acquire RTNL for every monitor pass (which
may occur as often as every 10 ms), the miimon has been converted to
do conditional locking.  A first pass inspects all slaves to determine
if any action is required, and if so, a second pass (after acquring RTNL)
is done to perform any actions (doing a complete rescan, as the situation
may have changed when all locks were released).
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

	Change locking in balance-rr transmit processing to use a free
running counter to determine which slave to transmit on.  Instead, a
free-running counter is maintained, and modulo arithmetic used to select
a slave for transmit.

	This removes lock operations from the TX path, and eliminates
a deadlock introduced by the conversion to work queues.
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

	Convert bonding timers to workqueues.  This converts the various
monitor functions to run in periodic work queues instead of timers.  This
patch introduces the framework and convers the calls, but does not resolve
various locking issues, and does not stand alone.
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Add missing &:

drivers/net/pasemi_mac.c: In function 'pasemi_mac_clean_rx':
drivers/net/pasemi_mac.c:553: warning: passing argument 1 of 'prefetch'
makes pointer from integer without a cast
Signed-off-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

 The dfx_bus_uninit() call is called from dfx_unregister() which is
__devexit and which is ultimately the ->remove call for the device.
Signed-off-by: NMaciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Fix off-by one in remove logic that just got introduced.
Signed-off-by: NStephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Plus minor formatting fixes.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Signed-off-by: NMatteo Croce <technoboy85@gmail.com>
Signed-off-by: NEugene Konev <ejka@imfi.kspu.ru>
Signed-off-by: NFelix Fietkau <nbd@openwrt.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Rather than hand-rolling our own prototype, make the code more
future-proof by using the standard irq_handler_t typedef.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Make irq handling more efficient, by passing board pointer via
request_irq() to our irq handler's dev_id argument.

This eliminates a table lookup upon each interrupt, and eliminates an
associated global variable (the table).
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

* pass card number to irq handler

* use card number in irq handler to avoid looping through each adapter
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

* invert sense of request_irq() test.  otherwise we will always fail,
  when IRQ is available.

* no need to use 'irq' function arg, its stored in a data struct already
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Polling currently calls the irq handler, which loops through all the
boards, calling the work function for all polling boards with work.

irq handling loops through all the boards, finding the specific board
that applies to us, and calling the work just for that one board.

The two logics are sufficiently different to warrant different
functions, rather than being slack and calling the same function in two
different ways.

This serves to make the interrupt handler a -lot- more efficient.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

No changes besides code movement and glue.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Remove always-false tests in irq handler.

Also a few other minor cleanups.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

None of the drivers with a struct pardevice's ->irq_func() hook ever
used the 'irq' argument passed to it, so remove it.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

parport_ieee1284_interrupt() was not using its first arg at all.
Delete.

parport_generic_irq()'s second arg makes its first arg completely
redundant.  Delete, and use port->irq in the one place where we actually
need it.

Also, s/__inline__/inline/ to make the code look nicer.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Several arches used the exact same code for their parport irq handling.
Make that code generic, in parport_irq_handler().

Also, s/__inline__/inline/ in include/linux/parport.h.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Driver for the watchdog timer. Still doesn't reboots the machine
on some boards, but we have improved and cleaned it
Signed-off-by: NMatteo Croce <technoboy85@gmail.com>
Signed-off-by: NNicolas Thill <nico@openwrt.org>
Signed-off-by: NEnrik Berkhan <Enrik.Berkhan@akk.org>
Signed-off-by: NChrister Weinigel <wingel@nano-system.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

ICP's Wafer 5823 SBC has, as far as I can tell, the same WDT as many,
if not all ICP's SBC's (that do have a WDT). I have tested it with
several boards, including Rocky 4783, Rocky 3703 and Rocky 3782. 

I propose a rename of the  Wafer 5823 watchdog timer driver
to something like "IPC (SBC) Watchdog Timer", to reflect that it
works with other IPC boards (maybe even all of them).
Signed-off-by: NVeljkovic Srdjan <sveljko@gvs.co.yu>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

The original serial-number calculations based on WWPN no longer
apply to newer ISPs (ISP24xx and ISP25xx).  These newer board's
serial number reside in the VPD.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

For recent ISPs, software during CS_UNDERRUN handling must
determine if the two residuals, firmware-calculated and FCP_RSP,
are different to recognize if a frame has been dropped.  Update
the driver to catch this condition, and clear the
SS_RESIDUAL_UNDER and lscsi_status bits.  This logic is
consistent with what earlier firmwares did by explicitly
cracking open the FCP_RSP statuses and clearing
SS_RESIDUAL_UNDER.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

Recent ISPs need only the single MMIO BAR to manipulate HW
registers.  Unfortunately, ISP21xx, ISP22xx, ISP23xx, and ISP63xx
type cards still require the I/O mapped region to manipulate the
FLASH via the two HW flash-registers (flash_address and
flash_data).
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

[SCSI] qla2xxx: Fix issue where final flash-segment updates were falling into the slow-path write handler.

Original implementation would not use the burst-write mechanisms
for requests equal to OPTROM_BURST_DWORDS transfer dwords.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

Since both NVRAM and VPD regions of the flash reside on unaligned
sector boundaries, during update, the driver must perform a
read-modify-write operation to the composite NVRAM/VPD region.
This affects ISP25xx type boards only.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

As the intermixing may cause issues where HCCR bits could be
cleared inappropriately during MSI/MSI-X interrupt handling.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@SteelEye.com>