e5318b53 ("ide: use the dma safe check for
REQ_TYPE_ATA_PC") introduced a regression which caused some ATAPI drives to
turn off DMA for REQ_TYPE_BLOCK_PC commands while burning and thus degrading
performance and ultimately causing an excessive amount of underruns.

The issue is documented also in:
http://bugzilla.kernel.org/show_bug.cgi?id=11742.
Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Tested-by: NValerio Passini <valerio.passini@unicam.it>
[bart: fixup patch description per comments from Sergei Shtylyov]
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

a) semicolon before the function body is a bad idea
b) it's const struct foo, not struct const foo
c) incidentally, it's ecard_remove_driver(), not ecard_unregister_driver()
d) compiling is occasionally useful.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Suggested-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

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

The "TEAC CD-ROM CD-224E" cannot close its tray, so add it to the NO_AUTOCLOSE
list. This will correct the "Can close tray" field of /proc/sys/dev/cdrom/info.
Signed-off-by: NMárton Németh <nm127@freemail.hu>
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
[bart: minor fixups to patch description]
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

This is the driver for the Toshiba TX4938 SoC EBUS controller ATA mode.
It has custom set_pio_mode and some hacks for big endian.
Signed-off-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: ralf@linux-mips.org
Cc: sshtylyov@ru.mvista.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

This is the driver for the Toshiba TX4939 SoC ATA controller.

This controller has standard ATA taskfile registers and DMA
command/status registers, but the register layout is swapped on big
endian.  There are some other endian issue and some special registers
which requires many custom dma_ops/tp_ops routines and build_dmatable.
Signed-off-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: ralf@linux-mips.org
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

based on suggestion from Sergei Shtylyov, there are two more places
where using pci_ioremap_bar() makes sense.
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Use the newly introduced pci_ioremap_bar() function in drivers/ide.
pci_ioremap_bar() just takes a pci device and a bar number, with the goal
of making it really hard to get wrong, while also having a central place
to stick sanity checks.
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Convert the driver to use ide_host_add() -- this seems to be a straightforward
change which I'm not sure why hasn't been done yet...

While doing this at last, get rid of:

- useless local copy of the 'sgiioc4_port_info' variable;

- unnecessary 'goto' and label...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: jeremy@sgi.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

The driver "forgot" to call iounmap() iff request_mem_region() call failed in
sgiioc4_ide_setup_pci_device()...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: jeremy@sgi.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Suggested-by: NRalf Baechle <ralf@linux-mips.org>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

To keep the size of changesets sane we split the switch by drivers;
to keep the damn thing bisectable we do the following:
	1) rename the affected methods, add ones with correct
prototypes, make (few) callers handle both.  That's this changeset.
	2) for each driver convert to new methods.  *ALL* drivers
are converted in this series.
	3) kill the old (renamed) methods.

Note that it _is_ a flagday; all in-tree drivers are converted and by the
end of this series no trace of old methods remain.  The only reason why
we do that this way is to keep the damn thing bisectable and allow per-driver
debugging if anything goes wrong.

New methods:
	open(bdev, mode)
	release(disk, mode)
	ioctl(bdev, mode, cmd, arg)		/* Called without BKL */
	compat_ioctl(bdev, mode, cmd, arg)
	locked_ioctl(bdev, mode, cmd, arg)	/* Called with BKL, legacy */
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

... convert to it in callers
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

store needed information in f_mode
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

commit 14c123f3 ("ide: cleanup
ide_build_dmatable()") accidentally reverted TRM290 fix introduced
by commit 22e05b45 ("ide-dma: fix
ide_build_dmatable() for TRM290").
Reported-by: NSergei Shtylylov <sshtylyov@ru.mvista.com>
Acked-by: NSergei Shtylylov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Fix the "unused variable" warning in init_hwif_scc() caused by the commit
48c3c107 (ide: add struct ide_host (take 3)).
Moreover, remove the write-only variable 'dma_status_port' in init_setup_scc()
about which gcc gives no warning and which has been there from the very start...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

By the time ide_dma_sgiioc4() gets called, sgiioc4_ide_setup_pci_device() will
have called ioremap() on the whole BAR0 region, so calling ioremap() on the DMA
registers means wasting a page. Replace this call by a mere address calculation,
based on the fact that IRQ registers (pointed to by 'hwif->io_ports.irq_addr')
are situated at offset 0 from BAR0.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: jeremy@sgi.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

The driver performs a number of checks on the virtual/physical addresses which
would always evaluate as true (except ide_dma_sgiioc4() -- always false):

- for sgiioc4_init_hwif_ports(), its caller, sgiioc4_ide_setup_pci_device(),
  guarantees that 'ctrl_port' and 'irq_port' parameters are never 0;

- in sgiioc4_read_status(), we always read the IDE status register, so there's
  no need to check the register's address (must be a leftover from the times
  when this function implemented the INB() method);

- in ide_dma_sgiioc4(), 'dma_base' can never be 0 as IOC4_DMA_OFFSET is not 0.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: jeremy@sgi.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* Factor out chipset initialization code from delkin_cb_probe()
  to delkin_cb_init_chipset().

* Assign ->init_chipset in struct delkin_cb_port_info.

* Add delkin_cb_{suspend,resume}() (->suspend/->resume methods).

Fixes kernel bugzilla bug #11735:

http://bugzilla.kernel.org/show_bug.cgi?id=11735

Tested-by: bumble.bee@xs4all.nl
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

No big loss since HPT343/363 controllers are properly supported
by pata_hpt3x3 driver from Alan Cox.
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

There should be no functional changes caused by this patch.
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Since the maskproc() method calls either mirror the interrupt en/disable via
the nIEN bit of the device control register done by the IDE core before issuing
a command or unmask the interrupt after a command executed in polled mode (when
interrupt is already not expected), it is pointless to implement this method
by manipulating the nIEN bit...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: jeremy@sgi.com
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Since the maskproc() method calls either mirror the interrupt en/disable via
the nIEN bit of the device control register done by the IDE core before issuing
a command or unmask the interrupt after a command executed in polled mode (when
interrupt is already not expected), it is pointless to manipulate the nIEN bit
in this method; therefore, just do nothing for the drives not on the quirk list.
Move the code to the left by using an early return and the 'else if' construct,
while at it....
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
[bart: fix checkpatch.pl warning]
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Apparently, there is no sense in unmasking IRQ on the controller when you call
disable_irq_nosync() before doing this, set the nIEN bit afterwards, and then
unmask IRQ again after the command completion, hence 0 passed to SELECT_MASK()
before issuing the command in ide_config_drive_speed() is probably just a typo.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Fixup for commit 1785192b
("hpt366: add hpt3xx_disable_fast_irq() helper"):

   CC      drivers/ide/pci/hpt366.o
drivers/ide/pci/hpt366.c: In function `init_hwif_hpt366':
drivers/ide/pci/hpt366.c:1290: warning: unused variable `dev'
Reported-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
[bart: split-up this change from a bigger patch]
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Add some more verbosity to key function calls in ide-cd debug code. While at it,
delete a superfluous comment.
Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* Add struct ide_disk_ops containing protocol specific methods.

* Add 'struct ide_disk_ops *' to ide_drive_t.

* Convert ide-{disk,floppy} drivers to use struct ide_disk_ops.

* Merge ide-{disk,floppy} drivers into generic ide-gd driver.

While at it:
- ide_disk_init_capacity() -> ide_disk_get_capacity()
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Turn ide_driver_t's 'proc' field into ->proc_entries method
(and also 'settings' field into ->proc_devsets method).  Then
update all device drivers accordingly.

There should be no functional changes caused by this patch.
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

- idefloppy_ref_mutex -> ide_disk_ref_mutex
- idefloppy_cleanup_obj() -> ide_disk_release()
- ide_floppy_get() -> ide_disk_get()
- ide_floppy_put() -> ide_disk_put()
- ide_floppy_capacity() -> ide_gd_capacity()
- ide_floppy_remove() -> ide_gd_remove()
- ide_floppy_probe() -> ide_gd_probe()
- idefloppy_driver -> ide_gd_driver
- idefloppy_open() -> ide_gd_open()
- idefloppy_release() -> ide_gd_release()
- idefloppy_getgeo() -> ide_gd_getgeo()
- idefloppy_media_changed() -> ide_gd_media_changed()
- idefloppy_revalidate_disk() -> ide_gd_revalidate_disk()
- idefloppy_ops -> ide_gd_ops
- idefloppy_init() -> ide_gd_init()
- idefloppy_exit() -> ide_gd_exit()

- 'floppy' -> 'idkp' in ide_disk_*() and ide_gd_*()
- idefloppy_floppy_t -> struct ide_floppy_obj

There should be no functional changes caused by this patch.
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

While at it:
- idefloppy_do_request() -> ide_floppy_do_request()
- idefloppy_end_request() -> ide_floppy_end_request()
- idefloppy_setup() -> ide_floppy_setup()

There should be no functional changes caused by this patch.
Acked-by: NBorislav Petkov <petkovbb@gmail.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>