sky2: transmit timeout

The transmit timeout code could hang, and it would not clear out
problems if the hardware was stuck.  Change the code to effectively do
a device down/up similar to the suspend/resume code.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

sky2: flow control negotiation for Yukon-FE

The Yukon-FE chip doesn't do gigabit and has a differen PHY internally.
On this chip, phy status register doesn't properly reflect the result
of flow control negotiation. To workaround the problem and avoid having
to have so much chip dependent code; compute the result of flow control
by looking at the local and remote advertised bits.

Signed-off-by: Stephen Hemmminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

sky2: no need to reset pause bits on shutdown

Resetting the pause bits on shutdown is not necessary.
The code was inherited from the vendor driver, and it is currently #ifdef'd
out there as well.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

sky2: dont flush good pause frames

Don't mark pause frames as errors. This problem caused transmitter not
to pause and would effectively take out a gigabit switch because the
it can't handle overrun.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

cxgb3 Fix copyrights in the cxgb3 driver.

Fix copyrights in the cxgb3 driver.

Remove the Open Grid Computing copyright.  It shouldn't be there.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

gianfar: don't duplicate gfar_error()

It was hardly necessary to repeat most of the code from gfar_error() in
gfar_interrupt(), especially having some inconsistencies between the two.
So, make the gfar_interrupt() just call gfar_error(), and not acknowledge
the interrupts itself as gfar_{receive/transmit/error}() do it anyway.
While at it, also clarify/cleanup debug messages in gfar_error()...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

NetXen: Fix to get the driver working after sparse changes

Signed-off-by: Amit S. Kale <amitkale@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

net/wan/pc300too.c: pci_module_init to pci_register_driver

Convert pci_module_init() to pci_register_driver().

Signed-off-by: Richard Knutsson <ricknu-0@student.ltu.se>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: bump version number

Bump the version number.

Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: add L1 device id to pci_ids, then use it

Add device id for the Attansic L1 chip to pci_ids.h, then use it.

Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: remove unused define

Remove unused define from atl1_main.c.

Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: read MAC address from register

On some Asus motherboards containing the L1 NIC, the MAC address is
written by the BIOS directly to the MAC register during POST, and is
not stored in eeprom.  If we don't succeed in fetching the MAC address
from eeprom or spi, try reading it directly from the MAC register.
Suggested by Xiong Huang.

And do some cleanup while we've got the hood up...

Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: fix bad ioread address

An ioread32 statement reads the wrong address.  Fix it.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

atl1: drop NET_PCI from Kconfig

The atl1 driver doesn't need NET_PCI.  Remove it from Kconfig.
Noticed by Chad Sprouse.

Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

PCMCIA-NETDEV : the 2nd argument of el3_interrrupt and smc_interrupt is struct net_device*

Signed-off-by: Komuro <komurojun-mbn@nifty.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

phylib: Add support for Marvell 88e1111S and 88e1145

Changes include:
* New support for 88e1145
* New support for 88e111s
* Fixing 88e1101 driver to not match non-88e1101 PHYs
* Increases in feature support across Marvell PHY product line
* Fixes a bunch of whitespace issues found by Lindent

Signed-off-by: Andrew Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

Merge branch 'master' into upstream

Merge branch 'gfar' of /linux/kernel/git/galak/powerpc into upstream

Merge AT91, EP93xx, General devel, PXA, S3C, V6+ and Xscale trees

[ARM] 4211/1: Provide a defconfig for ns9xxx

This is a follow up for the patch providing the base support for the
ns9xxx machine type.

Signed-off-by: Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4210/1: base for new machine type "NetSilicon NS9360"

Signed-off-by: Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4222/1: S3C2443: Remove reference to missing S3C2443_PM

S3C2443_PM is not defined in our Kconfig, so
remove the reference from CPU_S3C2443 to stop
the configuration process warning about it.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Documentation/kernel-docs.txt update.

Signed-off-by: James Nelson <james4765@gmail.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

arch/cris: typo in KERN_INFO

Typo in KERN_INFO.

Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Storage class should be before const qualifier

The C99 specification states in section 6.11.5:

The placement of a storage-class specifier other than at the
beginning of the declaration specifiers in a declaration is an
obsolescent feature.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

kernel/printk.c: comment fix

Signed-off-by: Patrick Pletscher <pat@pletscher.org>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

update I/O sched Kconfig help texts - CFQ is now default, not AS.

Change I/O scheduler description to correctly show CFQ as being the default
scheduler and not the anticipatory scheduler that previously was default.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Remove duplicate listing of Cris arch from README

Mikael Pettersson pointed out to me that a recent patch of mine (commit
620034c84d1d939717bdfbe02c51a3fee43541c3), that made some corrections to the
README file, accidentally listed the Cris architecture twice.  Whoops.

This patch removes the duplicate.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

kbuild: more doc. cleanups

Fix typos/spellos in kbuild/makefiles.txt.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

doc: make doc. for maxcpus= more visible

Some people are confused about maxcpus=1 and maxcpus=0,
so put the documentation text from init/main.c into
Documentation/kernel-parameters.txt also.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

drivers/net/eexpress.c: remove duplicate comment

Signed-off-by: Adrian Bunk <bunk@stusta.de>

add a help text for BLK_DEV_GENERIC

This fixes kernel Bugzilla #4933.

Signed-off-by: Adrian Bunk <bunk@stusta.de>

correct a dead URL in the IP_MULTICAST help text

Reported in kernel Bugzilla #6216.

Signed-off-by: Adrian Bunk <bunk@stusta.de>

fix the BAYCOM_SER_HDX help text

This fixes kernel Bugzilla #4076.

Signed-off-by: Adrian Bunk <bunk@stusta.de>

fix SCSI_SCAN_ASYNC help text

Signed-off-by: Adrian Bunk <bunk@stusta.de>

trivial documentation patch for platform.txt

Found a couple of typos in the Documentation/driver-model/platform.txt
file. This patch fixes both of them.

Signed-off-by: Erik Hovland <erik@hovland.org>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Fix typos concerning hierarchy

        heirarchical, hierachical -> hierarchical
        heirarchy, hierachy -> hierarchy

Signed-off-by: Uwe Kleine-König <zeisberg@informatik.uni-freiburg.de>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Fix comment typo "spin_lock_irqrestore".

Fix "spin_lock_irqrestore" to "spin_unlock_irqrestore."

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Fix misspellings of "agressive".

Fix the various misspellings of "agressive", as well as a couple
other things on the same lines while we're there.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

drivers/scsi/a100u2w.c: trivial typo patch

Trivial typo fix.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Correct trivial typo in log2.h.

Single typo correction in include/linux/log2.h.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Remove useless FIND_FIRST_BIT() macro from cardbus.c.

Delete the definition of the unused FIND_FIRST_BIT() macro.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Replace remaining references to "driverfs" with "sysfs".

Globally, s/driverfs/sysfs/g.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Use ARRAY_SIZE() macro in i386 relocs.c file

Change the explicit code in the relocs.c file to use ARRAY_SIZE()
and add a definition of ARRAY_SIZE() since this is a userspace program
and wouldn't include kernel.h.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Various typo fixes.

Correct mis-spellings of "algorithm", "appear", "consistent" and
(shame, shame) "kernel".

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

kbuild: Replace remaining "depends" with "depends on"

Replace the very few remaining "depends" Kconfig directives with
"depends on".

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

rio: typo in bitwise AND expression.

The line :

    hp->Mode &= !RIO_PCI_INT_ENABLE;

is obviously wrong as RIO_PCI_INT_ENABLE=0x04 and is used as a bitmask
2 lines before. Getting no IRQ would not disable RIO_PCI_INT_ENABLE
but rather RIO_PCI_BOOT_FROM_RAM which equals 0x01.

Obvious fix is to change ! for ~.

Signed-off-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Adrian Bunk <bunk@stusta.de>

Merge branch 'master' into 85xx

[POWERPC] 83xx: Renamed MPC8323 MDS dts and defconfig to match other boards

Renamed the MPC8323 MDS and defconfig to match the naming convention followed
by other MDS boards.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[POWERPC] 83xx: Updated and renamed MPC8360PB to MPC836x MDS

The MPC836x PB board is really just one part of the MPC836x MDS.  We currently
name all other PB boards as MDS.  Removed all references to PB and replaced
with MDS.  Additionally renamed the .dts to match the defconfig (mpc836x_mds*).

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[POWERPC] 83xx: Use of_platform_bus_probe to setup QE devices

Use of_platform_bus_probe to setup devices on the of_platform_bus since its
much cleaner.  We explicitly specify the bus ids since the we want to get rid
of the default mechanism in the future.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[POWERPC] 83xx: use default value of loops_per_jiffy

Use the default value setup by initialization of loops_per_jiffy, its close
enough for 83xx and will get fixed up by calibrate_delay().

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[POWERPC] 83xx: Remove obsolete setting of ROOT_DEV.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[POWERPC] 83xx: Cleaning up machine probing and board initcalls

Cleaned up the probing functionality to be more consistent across all 83xx
boards and added machine_is() protection around board initcalls to ensure they
only do something if we are actually running on that board.

Additionally, removed some dead code on mpc832x_mds.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

[ARM] 4221/1: S3C2443: DMA support

Add initialisation and mappings for S3C2443 DMA
system

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4220/1: S3C24XX: DMA system initialised from sysdev

This patch gets the DMA system for the S3C24XX
ready for the S3C2443, which requires 6 dma channels
at a different stride, and different base IRQ.

The DMA system is now initialised from the same
drivers which apply the DMA mappings, as well
as removing the DMA sysdev intialisation out of
the main init code (which is now being called
from a sysdev probe, so cannot add a new sysdev)

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4219/1: S3C2443: DMA source definitions

Defines for the S3C2443 DMA source selection,
and update the maximum channels to 6 if the S3C2443
is selected.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4218/1: S3C2412: fix CONFIG_CPU_S3C2412_ONLY wrt to S3C2443

Update the Kconfig of arch/arm/mach-s3c2443 to
add the CONFIG_CPU_S3C2443 to the list of config
variabls which mean CONFIG_CPU_S3C2412_ONLY cannot
be set

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4217/1: S3C24XX: remove the dma channel show at startup

Remove the DMA code's channel printing at startup
as this is firstly a waste of console output on
initialsaion, and secondly is going to be obsolete
once the S3C2443 DMA code has been merged

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Merge branch 'master' into 83xx

ieee1394: fix another deadlock in nodemgr

A "modprobe ohci1394; sleep 1.5; modprobe -r ohci1394" could get stuck
in uninterruptible state, especially if an external node was connected.
http://bugzilla.kernel.org/show_bug.cgi?id=7792

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>

ieee1394: cycle timer read extension for raw1394

This implements the simultaneous read of the isochronous cycle timer and
the system clock (in usecs).  This allows to express the exact receive
time of an ISO packet as a system time with microsecond accuracy.
http://bugzilla.kernel.org/show_bug.cgi?id=7773

The counterpart patch for libraw1394 can be found at
http://thread.gmane.org/gmane.linux.kernel.firewire.devel/8934

Patch update (Stefan R.):  Disable preemption and local interrupts.
Prevent integer overflow.  Add paranoid error checks and kerneldoc to
hpsb_read_cycle_timer.  Move it to other ieee1394_core high-level API
functions.  Change comments.  Adjust whitespace.  Rename struct
_raw1394_cycle_timer.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Pieter Palmers <pieterp@joow.be>
Acked-by: Dan Dennedy <dan@dennedy.org>

[ARM] 4090/2: avoid clash between PXA and SA1111 defines

The Lubbock platform uses both a PXA25x and a SA1111 at the same time.
Both chips have the same "Serial Audio Controller" registers although
the SA1111 one is never expected to be used in preference to the PXA25x
one. So let's disable the SA1111 defines whenever compilation is for a
PXA architecture and make the PXA defines always defined.

This removes a bunch of "already defined" warnings as well since the
current hack to prevent them depended on include ordering which wasn't
always right.

While at it, clean up the SA1111 defines allowing to get rid of the
__CCREG() macro.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4216/1: add .gitignore entries for ARM specific files

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4214/2: S3C2410: Add Armzone QT2410

Patch from: Harald Welte <laforge@openmoko.org>

Add support for the Armzone QT2410 system, with
basic peripheral support for TFT display, SPI
and LEDs.

Signed-off-by: Harald Welte <laforge@openmoko.org>
Signed-off-by: Ben Dooks <ben-linux@fluf.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4215/1: s3c2410 usb device:  per-platform vbus_draw

Introduce a platform_device (machine) specific callback function
which gets called when the amount of power we can draw from Vbus
has changed.

Signed-off-by: Harald Welte <laforge@openmoko.org>
Signed-off-by: Arnaud Patard <arnaud.patard@rtp-net.org>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4213/1: S3C2410 - Update definition of ADCTSC_XY_PST

Update S3C2410_ADCTSC_XY_PST macro to allow setting the ADCTSC_XY_PST
bits.

Signed-off-by: Arnaud Patard <arnaud.patard@rtp-net.org>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[POWERPC] Dispose irq mapping when done in mpc52xx_serial.c

Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Paul Mackerras <paulus@samba.org>

Pull bugzilla-7897 into release branch

Pull bugzilla-7887 into release branch

Pull bugzilla-7570 into release branch

Pull bugzilla-7200 into release branch

Pull bugzilla-7122 into release branch

Pull bugzilla-5534 into release branch

Pull remove-hotkey into release branch

Pull sony into release branch

Pull fluff into release branch

Conflicts:

arch/x86_64/pci/mmconfig.c
drivers/acpi/bay.c

Signed-off-by: Len Brown <len.brown@intel.com>

ACPI: IA64: react to acpi_table_parse() return value change

acpi_boot_init() is making a bad check on the return
status from acpi_table_parse(). acpi_table_parse() now
returns zero on success, one on failure.

Signed-off-by: Aaron Young <ayoung@sgi.com>
Signed-off-by: Len Brown <len.brown@intel.com>

ACPI: hotkey: remove driver, per feature-removal-schedule.txt

Signed-off-by: Len Brown <len.brown@intel.com>

ide: make ide_hwif_t.ide_dma_host_on void (v2)

* since ide_hwif_t.ide_dma_host_on is called either when drive->using_dma == 1
  or when return value is discarded make it void, also drop "ide_" prefix
* make __ide_dma_host_on() void and drop "__" prefix

v2:
* while at it rename atiixp_ide_dma_host_on() to atiixp_dma_host_on()
  and sgiioc4_ide_dma_host_on() to sgiioc4_dma_host_on().
  [ Noticed by Sergei Shtylyov <sshtylyov@ru.mvista.com>. ]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: make ide_hwif_t.ide_dma_{host_off,off_quietly} void (v2)

* since ide_hwif_t.ide_dma_{host_off,off_quietly} always return '0'
  make these functions void and while at it drop "ide_" prefix
* fix comment for __ide_dma_off_quietly()
* make __ide_dma_{host_off,off_quietly,off}() void and drop "__" prefix

v2:
* while at it rename atiixp_ide_dma_host_off() to atiixp_dma_host_off(),
  sgiioc4_ide_dma_{host_off,off_quietly}() to sgiioc4_dma_{host_off,off_quietly}()
  and sl82c105_ide_dma_off_quietly() to sl82c105_dma_off_quietly()
  [ Noticed by Sergei Shtylyov <sshtylyov@ru.mvista.com>. ]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: add ide_set_dma() helper (v2)

* add ide_set_dma() helper and make ide_hwif_t.ide_dma_check return
  -1 when DMA needs to be disabled (== need to call ->ide_dma_off_quietly)
   0 when DMA needs to be enabled  (== need to call ->ide_dma_on)
   1 when DMA setting shouldn't be changed
* fix IDE code to use ide_set_dma() instead if using ->ide_dma_check directly

v2:
* updated for scc_pata

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

sgiioc4: fix sgiioc4_ide_dma_check() to enable/disable DMA properly

* use sgiioc4_ide_dma_{on,off_quietly}() instead of changing
  drive->using_dma directly
* fix warning message
* add FIXME

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: disable DMA in ->ide_dma_check for "no IORDY" case (v2)

If DMA is unsupported ->ide_dma_check should disable DMA.

v2:
* updated for scc_pata

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: convert ide_hwif_t.mmio into flag (v2)

All users of ->mmio == 1 are gone so convert ->mmio into flag.

Noticed by Alan Cox.

v2:
* updated for scc_pata

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: use PIO/MMIO operations directly where possible (v2)

This results in smaller/faster/simpler code and allows future optimizations.
Also remove no longer needed ide[_mm]_{inl,outl}() and ide_hwif_t.{INL,OUTL}.

v2:
* updated for scc_pata

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: add ide_use_fast_pio() helper (v3)

* add ide_use_fast_pio() helper for use by host drivers

* add DMA capability and hwif->autodma checks to ide_use_dma()

  - au1xxx-ide/it8213/it821x drivers didn't check for (id->capability & 1)

    [ for the IT8211/2 in SMART mode this check shouldn't be made but since
      in it821x_fixups() we set DMA bit explicitly:

               if(strstr(id->model, "Integrated Technology Express")) {
                       /* In raid mode the ident block is slightly buggy
                          We need to set the bits so that the IDE layer knows
                          LBA28. LBA48 and DMA ar valid */
                       id->capability |= 3;            /* LBA28, DMA */

       we are better off using generic helper if we can ]

  - ide-cris driver didn't set ->autodma

    [ before the patch hwif->autodma was only checked in the chipset specific
      hwif->ide_dma_check implementations, for ide-cris it is cris_dma_check()
      function so there no behavior change here ]

v2:
* updated patch description (thanks to Alan Cox for the feedback)

v3:
* updated for scc_pata driver

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: unexport ide_set_xfer_rate() (v2)

In cmd64x, siimage and scc_pata drivers:
* don't set drive->init_speed as it should be already
  set by successful execution of ide_set_xfer_rate()
* use hwif->speedproc functions directly

Above changes allows removal of EXPORT_SYMBOL_GPL(ide_set_xfer_rate).

v2:
* updated for scc_pata driver

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: remove ide_drive_t.usage

This field is no longer used by the core IDE code so fix ide-{disk,floppy}
drivers to keep openers count in the driver specific objects and remove
it from ide-{cd,scsi,tape} drivers (it was write-only).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: remove ide_pci_device_t tables with only one entry

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: remove write-only ide_hwif_t.no_dsc flag

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide: remove write-only ide_pio_data_t.blacklisted

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

sis5513: sis5513_config_xfer_rate() cleanup

* remove bogus comment for sis5513_config_xfer_rate()
* there is no need to call config_drive_art_rwp() because
  it is called by config_art_rwp_pio()
* remove needless wrapper
* remove stale "TODO" comment
  (IDE core should provide generic tuning code)

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

piix: cleanup

* disable DMA masks if no_piix_dma is set and remove now
  not needed no_piix_dma_check from piix_config_drive_for_dma()
* there is no need to read register 0x55 in init_hwif_piix()
* move cable detection code to piix_cable_detect()
* remove unreachable 82371MX code from init_hwif_piix()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

svwks: small cleanup

* remove redundant svwks_ide_dma_end() [ __ide_dma_end() is used by default ]
* remove init_dma_svwks() so the default ide_setup_dma() function is used
  [ init_setup_csb6() takes care of not initializing disabled channels ]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

cs5530: small cleanup

* BUG() on unknown DMA mode in cs5530_config_dma()
* there is no need to call hwif->ide_dma_host_{off,on}() in
  cs5530_config_dma() because hwif->ide_dma_host_{off,on}()
  is called by hwif->ide_dma_off_{quietly,on}()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

hpt366: remove redundant check from init_dma_hpt366()

->init_dma() cannot be called with dmabase == 0
(see drivers/ide/setup-pci.c)

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

trm290: remove redundant CONFIG_BLK_DEV_IDEDMA #ifdef-s

In drivers/ide/Kconfig BLK_DEV_TRM290 depends on
BLK_DEV_IDEDMA_PCI (on which is BLK_DEV_IDEDMA dependant on).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

au1xxx-ide: remove dead code

'speed' is always equal to 'mode' when ide_config_drive_speed() is called

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

pdc202xx_old: remove dead code

CONFIG_PDC202XX_MASTER config option doesn't exist

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>