* Add IDE_HFLAG_ABUSE_{PREFETCH,FAST_DEVSEL,DMA_MODES} flags
  and set them in ht6560, cmd640, cmd64x and sc1200 host drivers.

* Add set_pio_mode_abuse() for checking if host driver has a non-standard
  ->tuneproc() implementation and use it in do_special().

* Add ide_set_pio() for setting PIO mode (it uses hwif->pio_mask to find
  the maximum PIO mode supported by the host), also add ide_set_max_pio()
  wrapper for ide_set_pio() to use for auto-tuning.  Convert users of
  ->tuneproc to use ide_set{_max}_pio() where possible.  This leaves only
  do_special(), set_using_pio(), ide_hwif_restore() and ide_set_pio() as
  a direct users of ->tuneproc.

* Remove no longer needed ide_get_best_pio_mode() calls and printk-s
  reporting PIO mode selected from ->tuneproc implementations.

* Rename ->tuneproc hook to ->set_pio_mode and make 'pio' argument const.

* Remove stale comment from ide_config_drive_speed().

v2:
* Fix "ata_" prefix (Noticed by Jeff).

v3:
* Minor cleanups/fixups per Sergei's suggestions.

v4:
* Fix compile problem in drivers/ide/pci/cmd640.c
  (Noticed by Andrew Morton).

* Improve some ->set_pio_mode comments.
Reviewed-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Some of the code has been gradually transitioned to using the proper
struct request_queue, but there's lots left. So do a full sweet of
the kernel and get rid of this typedef and replace its uses with
the proper type.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

* Do the same thing as probe_hwif() and always disable DMA so chipset DMA
  enabled bit gets cleared (if the drive doesn't support DMA ide_set_dma()
  won't try to tune it anyway).

* Add TODO comment about respecting ->using_dma setting.
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide-disk calls

        ide_end_request(drive, 0, 0);

to finish an unknown request, but this doesn't work so well for non-fs
requests, since ide_end_request() internally looks at ->hard_cur_sectors
to see how much data to end. Only file system requests store a transfer
value in there, pc requests fill out ->data_len as a byte based transfer
value instead.

Since we ask to end 0 bytes of that request, it will never be terminated
and ide-disk gets stuck in a loop "handling" that same request over and
over.

Switch __ide_end_request() to take a byte based transfer count, and
adjust ide_end_request() to look at the right field to determine how
much IO to end when it's being passed in 0.
Acked-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Tested-By: NGiacomo Catenazzi <cate@debian.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since ide_dma_timeout() method's result is discarded, make it return 'void'.
While at it, drop 'ide_' from the method's name, drop the '__' prefix from
the default method's name, and do some cleanups in this method driver-wise:

- in ide-dma.c, au1xxx-ide.c, and pdc202xx_old.c, define/use 'hwif' variable;

- in au1xxx-ide.c, get rid of commented out printk();

- in sl82c105.c, get rid of unnecessary variables.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Since ide_dma_lostirq() method's result is discarded, make it return 'void'.
While at it, rename the method to dma_lost_irq(), drop the '__' prefix from the
default method's name, and do some cleanups in this method driver-wise:

- in aec62xx.c, rename the method in accordance with other drivers, and get rid
  of unnecessary variables there;

- in pdc202xx_old.c, define/use 'hwif' variable;

- in sgiioc4.c, rearrange the code to call the resetproc() method directly.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide_start_power_step() and set_using_dma() were missing ->dma_off_quietly
call (comment in probe_hwif() states that DMA should be always cleared before
tuning is attempted).  Fix it.
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

PIO should be restored also for ATAPI devices during resume, fix it.
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

It is possible for the timer expiry function to run even though the
request has already been handled: ide_timer_expiry() only checks that
the handler is not NULL, but it is possible that we have handled a
request (thus clearing the handler) and then started a new request
(thus starting the timer again, and setting a handler). 

A simple way to exhibit this is to set the DMA timeout to 1 jiffy and
run dd: The kernel will panic after a few minutes because
ide_timer_expiry() tries to add a timer when it's already active.

To fix this, we simply add a request generation count that gets
incremented at every interrupt, and check in ide_timer_expiry() that
we have not already handled a new interrupt before running the expiry
function.
Signed-off-by: NSuleiman Souhlal <suleiman@google.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

IDE error recovery is using IDLE IMMEDIATE if the drive is busy or has DRQ set.
This violates the ATA spec (can only send IDLE IMMEDIATE when drive is not
busy) and really hoses up some drives (modern drives will not be able to
recover using this error handling).  The correct thing to do is issue a SRST
followed by a SET FEATURES command.  This is what Western Digital recommends
for error recovery and what Western Digital says Windows does.  It also does
not violate the ATA spec as far as I can tell.

Bart:
* port the patch over the current tree
* undo the recalibration code removal
* send SET FEATURES command after checking for good drive status
* don't check whether the current request is of REQ_TYPE_ATA_{CMD,TASK}
  type because we need to send SET FEATURES before handling any requests
* some pre-ATA4 drives require INITIALIZE DEVICE PARAMETERS command before
  other commands (except IDENTIFY) so send SET FEATURES only if there are
  no pending drive->special requests
* update comments and patch description
* any bugs introduced by this patch are mine and not Suleiman's :-)
Signed-off-by: NSuleiman Souhlal <suleiman@google.com>
Acked-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* 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: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* 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: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

patch 1/2 (revised):
- Fix drive->waiting_for_dma to work with CDB-intr devices.
- Do the dma status clearing in ide_intr() and add a new
  hwif->ide_dma_clear_irq for Intel ICHx controllers.

Revised per Alan, Sergei and Bart's advice.

Patch against 2.6.20-rc6. Tested ok on my ICH4 and pdc20275 adapters.
Please review/apply, thanks.
Signed-off-by: NAlbert Lee <albertcc@tw.ibm.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: "Adam W. Hawks" <awhawks@us.ibm.com>
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.

The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around.  On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).

Where appropriate, an arch may override the generic storage facility and do
something different with the variable.  On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.

Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions.  Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller.  A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.

I've build this code with allyesconfig for x86_64 and i386.  I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.

This will affect all archs.  Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:

	struct pt_regs *old_regs = set_irq_regs(regs);

And put the old one back at the end:

	set_irq_regs(old_regs);

Don't pass regs through to generic_handle_irq() or __do_IRQ().

In timer_interrupt(), this sort of change will be necessary:

	-	update_process_times(user_mode(regs));
	-	profile_tick(CPU_PROFILING, regs);
	+	update_process_times(user_mode(get_irq_regs()));
	+	profile_tick(CPU_PROFILING);

I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().

Some notes on the interrupt handling in the drivers:

 (*) input_dev() is now gone entirely.  The regs pointer is no longer stored in
     the input_dev struct.

 (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking.  It does
     something different depending on whether it's been supplied with a regs
     pointer or not.

 (*) Various IRQ handler function pointers have been moved to type
     irq_handler_t.
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)

Add a step to the IDE PM state machine that reprograms disk PIO timings
as the first step on resume. This prevents ide deadlock on
resume-from-ram on my nforce3-based laptop.

An earlier implementation was written entirely within the amd74xx ide
driver, but Alan helpfully pointed out that this is the correct thing to
do globally. Still, I'm only calling hwif->tuneproc() for disks, based
on two things:

 - The existing state machine is already passed over for non-disk drives
 - Previous testing on my laptop shows that the hangs are related only
   to the disk - suspend/resume from a livecd showed that there's no
   need for this on the cdrom.
Signed-off-by: NJason Lunz <lunz@falooley.org>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Brad Campbell <brad@wasp.net.au>
Cc: David Brownell <david-b@pacbell.net>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In 2.6.15.1 I encountered some IDE crashes when unplugging IDE cables to
emulate disk errors.  Below is a patch against 2.6.16 which I think still
applies.

1.  The first BUG_ON could trigger when a PREFLUSH IO fails (it would
   fail the original barrier request which hasn't been marked REQ_STARTED
   yet).

2. the rq could have been dequeued already (same as 1).

3.  HWGROUP(drive)->rq could be NULL because of the ide_error() several
   lines earlier.
Signed-off-by: NHua Zhong <hzhong@gmail.com>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

After Christophs SCSI change, the only usage left is RQ_ACTIVE
and RQ_INACTIVE. The block layer sets RQ_INACTIVE right before freeing
the request, so any check for RQ_INACTIVE in a driver is a bug and
indicates use-after-free.

So kill/clean the remaining users, straight forward.
Signed-off-by: NJens Axboe <axboe@suse.de>

As the comments indicates in blkdev.h, we can fold it into ->end_io_data
usage as that is really what ->waiting is. Fixup the users of
blk_end_sync_rq().
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Right now ->flags is a bit of a mess: some are request types, and
others are just modifiers. Clean this up by splitting it into
->cmd_type and ->cmd_flags. This allows introduction of generic
Linux block message types, useful for sending generic Linux commands
to block devices.
Signed-off-by: NJens Axboe <axboe@suse.de>

lockdep needs to have the waitqueue lock initialized for on-stack waitqueues
implicitly initialized by DECLARE_COMPLETION().  Annotate on-stack completions
accordingly.

Has no effect on non-lockdep kernels.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Make use of local_irq_enable_in_hardirq() API to annotate places that enable
hardirqs in hardirq context.

Has no effect on non-lockdep kernels.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NJörn Engel <joern@wohnheim.fh-wedel.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

If the controller FIFO cleared automatically on error we must not try
and drain it as this will hang some chips.

Based in concept on a broken patch from -mm some while back
Signed-off-by: NAlan Cox <alan@redhat.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

During an STR resume cycle, the ide master disk times-out when there is
also a slave present (especially CD).  Increasing the timeout in ide-io
from 10,000 to 100,000 fixes this problem.
Acked-by: NAlan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

acquired (aquired)
contiguous (contigious)
successful (succesful, succesfull)
surprise (suprise)
whether (weather)
some other misspellings
Signed-off-by: NAndreas Mohr <andi@lisas.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

This is a patch from Alan that fixes a real ide-cd.c regression causing
bogus "Media Check" failures for perfectly valid Fedora install ISOs, on
certain CD-ROM drives.

This is a forward port to 2.6.16 (from RHEL) of the minimal changes for the
end of media problem.  It may not be sufficient for some controllers
(promise notably) and it does not touch the locking so the error path
locking is as horked as in mainstream.

From: Ingo Molnar <mingo@elte.hu>

I have ported the patch to 2.6.17-rc4 and tested it by provoking
end-of-media IO errors with an unaligned ISO image.  Unlike the vanilla
kernel, the patched kernel interpreted the error condition correctly with
512 byte granularity:

 hdc: command error: status=0x51 { DriveReady SeekComplete Error }
 hdc: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
 ide: failed opcode was: unknown
 ATAPI device hdc:
   Error: Illegal request -- (Sense key=0x05)
   Illegal mode for this track or incompatible medium -- (asc=0x64, ascq=0x00)
   The failed "Read 10" packet command was:
   "28 00 00 04 fb 78 00 00 06 00 00 00 00 00 00 00 "
 end_request: I/O error, dev hdc, sector 1306080
 Buffer I/O error on device hdc, logical block 163260
 Buffer I/O error on device hdc, logical block 163261
 Buffer I/O error on device hdc, logical block 163262

the unpatched kernel produces an incorrect error dump:

 hdc: command error: status=0x51 { DriveReady SeekComplete Error }
 hdc: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
 ide: failed opcode was: unknown
 end_request: I/O error, dev hdc, sector 1306080
 Buffer I/O error on device hdc, logical block 163260
 hdc: command error: status=0x51 { DriveReady SeekComplete Error }
 hdc: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
 ide: failed opcode was: unknown
 end_request: I/O error, dev hdc, sector 1306088
 Buffer I/O error on device hdc, logical block 163261
 hdc: command error: status=0x51 { DriveReady SeekComplete Error }
 hdc: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
 ide: failed opcode was: unknown
 end_request: I/O error, dev hdc, sector 1306096
 Buffer I/O error on device hdc, logical block 163262

I do not have the right type of CD-ROM drive to reproduce the end-of-media
data corruption bug myself, but this same patch in RHEL solved it.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: Jens Axboe <axboe@suse.de>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The IDE power management can just use the ->end_io_data member to store
it's data.
Signed-off-by: NJens Axboe <axboe@suse.de>

Since there's no longer any external user, we can make __ide_end_request()
static.
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Acked-by: NBartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There's a problem with the REQ_BLOCK_PC handling as well (bad ->data_len
handling) where it could actually complete a request ahead of time.  I
suggest we just back this out for now, I will resubmit it later when I'm
fully confident in it.

This reverts commit 8672d571Signed-off-by: NJens Axboe <axboe@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

To preserve the ->errors values for requests that failed, use the normal
completion path for that.
Signed-off-by: NJens Axboe <axboe@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch makes IDE use the new blk_complete_request() interface.
There's still room for improvement, as __ide_end_request() really
could drop the lock after getting HWGROUP->rq (why does it need to
hold it in the first place? If ->rq access isn't serialized, we are
screwed anyways).
Signed-off-by: NJens Axboe <axboe@suse.de>

Update IDE to use new blk_ordered.  This change makes the
following behavior changes.

* Partial completion of the barrier request is handled as
  failure of the whole ordered sequence.  No more partial
  completion for barrier requests.

* Any failure of pre or post flush request results in failure
  of the whole ordered sequence.

So, successfully completed ordered sequence guarantees that
all requests prior to the barrier made to physical medium and,
then, the while barrier request made to the physical medium.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJens Axboe <axboe@suse.de>

add @uptodate argument to end_that_request_last() and @error
to rq_end_io_fn().  there's no generic way to pass error code
to request completion function, making generic error handling
of non-fs request difficult (rq->errors is driver-specific and
each driver uses it differently).  this patch adds @uptodate
to end_that_request_last() and @error to rq_end_io_fn().

for fs requests, this doesn't really matter, so just using the
same uptodate argument used in the last call to
end_that_request_first() should suffice.  imho, this can also
help the generic command-carrying request jens is working on.
Signed-off-by: Ntejun heo <htejun@gmail.com>
Signed-Off-By: NJens Axboe <axboe@suse.de>

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

The logic in ide_do_request() doesn't guarantee that both drives will be
serviced after a call.  It may "forget" to service one in some
circumstances, including when one of the drive is suspended (it will
eventually fail to service the slave when the master is suspended for
example).  This prevents the wakeup requests that gets queued on wakeup
from sleep from beeing serviced in some cases when 2 drives are sharing
an IDE bus.

The problem is deep enough in the way this code works (and there are
probably a few other problematic but rare corner cases) and fixing it
would require some major rethinking of the way IDE decides which channel
to service.  This is not 2.6.14 material.  However, in the meantime,
Bart has accepted this simple workaround that will fix the crash on
wakeup from sleep since this specific corner case is actually hitting
users to get into 2.6.14.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch contains the most trivial from Rusty's trivial patches:
- spelling fixes
- remove duplicate includes
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Stop the disks from spinning down and up on suspend.
Signed-off-by: NMichal Schmidt <xschmi00@stud.feec.vutbr.cz>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!