ata: ahci: refactor init functions ...

Fold ahci_host_init and __ahci_host_init into ahci_add_host

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: ahci: implement init callback ...

Now that we have framework support for delayed probing of
ata devices implement it in the ahci driver to actually make use
of it.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Add init callback to ata_port_operations ...

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: add ata logical device to defer probe ...

ata device usually take a long time to spin up, so it makes sense
to only spend this time when the device is actually used.

This adds a logical ata device and attaches a 'probe' parameter
to it, similar to what MMC does.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: ahci: fix wrong link check ...

The following commit accidently bailed out if the link is up, not
when it's down as stated in the commit message.

| commit a3f74d9
| Author: Rob Herring <rob.herring@calxeda.com>
| Date:   Mon Feb 11 18:02:00 2013 +0100
|
|     ahci: handle COMINIT received during spin-up
|
|     Some Intel SSDs can send a COMINIT after the initial COMRESET. This causes
|     the link to go down and we need to re-initialize the link.
|
|     Signed-off-by: Rob Herring <rob.herring@calxeda.com>
|     Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
|     Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
|

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/misc'

ahci-generic: add oftree support ...

calxeda use the generic driver but have it's own compatible cf Linux kernel
Documentation

Cc: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ahci: handle COMINIT received during spin-up ...

Some Intel SSDs can send a COMINIT after the initial COMRESET. This causes
the link to go down and we need to re-initialize the link.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Use new device_platform_driver() macro for drivers ...

Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

sata-imx: fix depends on ARCH_IMX ...

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata sff: set device pointer in ata port ...

This was forgotten in the last ata updates and leads to NULL
pointer dereferences.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'kconfig'

Remove bad Kconfig reference to DISK_DRIVE ...

Barebox not contain symbol DISK_DRIVE, so remove all references to it.

Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Cleanup Kconfig files ...

This patch provides a global cleanup barebox Kconfig files. This includes
replacing spaces to tabs, formatting in accordance format, removing
extraneous lines and spaces. No functional changes.

Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Add i.MX AHCI driver ...

This adds the glue code for the i.MX SATA controller. This controller
needs some i.MX specific setup and some SoC specific setting outside
the controller itself. The code for setting up the correct clock source
for the SATA phy has been taken from U-Boot.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Add ahci support ...

This adds ahci controller support based on U-Boot ahci support. Unlike
U-Boot we do not push the SCSI layer in between, but use the ata interface
directly. Tested on a Freescale i.MX53.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Use dma_alloc for buffer ...

ATA devices using DMA may need a sufficiently aligned buffer, so use
dma_alloc instead of regular malloc.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: align ata command defines with kernel ...

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: split ide sff suport to separate file ...

Currently we only support oldschool IDE SFF devices. This is done
by registering a register layout struct and everything else is done
by the generic IDE SFF driver. Since modern ATA devices still use
ATA, but not the SFF interface anymore, split out the IDE SFF support
to a separate file to allow for other types of ata interfaces.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: fix status flags ...

Some status flags are wrong, fix them. All of them are currently unused,
so no functional change included.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: register disks as /dev/ata* ...

Using /dev/disk* for every type of device is not helpful. It increases
the chance that the user doesn't know which file corresponds to which
device. So rename ata device files to /dev/ata*. Also add a dev_info
about which device just has been registered.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Add i.MX PATA driver support ...

Straight forward driver, we only have to configure some additional
bits and then use the generic ide support.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ata: Allow partitions > 4GiB ...

Since barebox now uses 64bit for device accesses this is no
issue anymore.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

switch all platform_bus device/driver registering to platform_driver/device_register ...

now register_driver and register_device are for bus only usage.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>

Treewide: remove address of the Free Software Foundation ...

The FSF address has changed in the past. Instead of updating it
each time the address changes, just drop it completely treewide.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Change byte order detection mechanism to kernel style ...

The Linux Kernel defines only one of __LITTLE_ENDIAN and
__BIG_ENDIAN. Endianess can then be tested with #ifdef __xx_ENDIAN. Userspace
always defined both __LITTLE_ENDIAN and __BIG_ENDIAN and byteorder can then
be tested with #if __BYTE_ORDER == __xx_ENDIAN.

As we tend to use a lot of Kernel code in barebox we switch to use the kernel
way of determing the byte order.

As this always causes a lot of confusion add a check to include/common.h to
make sure only one of __LITTLE_ENDIAN and __BIG_ENDIAN is defined.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Allow use platform_ide driver without ALT-registers ...

Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Add driver for IDE like interfaces ...

This simple driver enables a generic driver for ATA type of devices to get
access to the so called 'register file' of an ATA drive.

Signed-off-by: Juergen Beisert <jbe@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ATA Disk Support: Add support for native ATA type drives ...

This changed patch removes more of the u-boot like code and replace it with
kernel like code.

commit 2a89669
Author: Juergen Beisert <jbe@pengutronix.de>
Date:   Fri Sep 30 15:06:26 2011 +0200

    ATA Disk Support: Add support for native ATA type drives

    Signed-off-by: Juergen Beisert <jbe@pengutronix.de>

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Remove 'disk_drive.c' as it is now replaced by generic partition handling ...

Signed-off-by: Juergen Beisert <jbe@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>