ATA_EHI_RESUME_LINK has two functions - promote reset to hardreset if
ATA_LFLAG_HRST_TO_RESUME is set and preventing EH from shortcutting
reset action when probing is requested.  The former is gone now and
the latter can easily be achieved by making EH to perform at least one
reset if reset is requested, which also makes more sense than
depending on RESUME_LINK flag.

As ATA_EHI_RESUME_LINK was the only EHI reset modifier, this also
kills reset modifier handling.
Signed-off-by: NTejun Heo <htejun@gmail.com>

When both soft and hard resets are available, libata preferred
softreset till now.  The logic behind it was to be softer to devices;
however, this doesn't really help much.  Rationales for the change:

* BIOS may freeze lock certain things during boot and softreset can't
  unlock those.  This by itself is okay but during operation PHY event
  or other error conditions can trigger hardreset and the device may
  end up with different configuration.

  For example, after a hardreset, previously unlockable HPA can be
  unlocked resulting in different device size and thus revalidation
  failure.  Similar condition can occur during or after resume.

* Certain ATAPI devices require hardreset to recover after certain
  error conditions.  On PATA, this is done by issuing the DEVICE RESET
  command.  On SATA, COMRESET has equivalent effect.  The problem is
  that DEVICE RESET needs its own execution protocol.

  For SFF controllers with bare TF access, it can be easily
  implemented but more advanced controllers (e.g. ahci and sata_sil24)
  require specialized implementations.  Simply using hardreset solves
  the problem nicely.

* COMRESET initialization sequence is the norm in SATA land and many
  SATA devices don't work properly if only SRST is used.  For example,
  some PMPs behave this way and libata works around by always issuing
  hardreset if the host supports PMP.

  Like the above example, libata has developed a number of mechanisms
  aiming to promote softreset to hardreset if softreset is not going
  to work.  This approach is time consuming and error prone.

  Also, note that, dependingon how you read the specs, it could be
  argued that PMP fan-out ports require COMRESET to start operation.
  In fact, all the PMPs on the market except one don't work properly
  if COMRESET is not issued to fan-out ports after PMP reset.

* COMRESET is an integral part of SATA connection and any working
  device should be able to handle COMRESET properly.  After all, it's
  the way to signal hardreset during reboot.  This is the most used
  and recommended (at least by the ahci spec) method of resetting
  devices.

So, this patch makes libata prefer hardreset over softreset by making
the following changes.

* Rename ATA_EH_RESET_MASK to ATA_EH_RESET and use it whereever
  ATA_EH_{SOFT|HARD}RESET used to be used.  ATA_EH_{SOFT|HARD}RESET is
  now only used to tell prereset whether soft or hard reset will be
  issued.

* Strip out now unneeded promote-to-hardreset logics from
  ata_eh_reset(), ata_std_prereset(), sata_pmp_std_prereset() and
  other places.
Signed-off-by: NTejun Heo <htejun@gmail.com>

Implement ata_qc_raw_nbytes() which determines the raw user-requested
size of a PC command.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Block layer alignment was used for two different purposes - memory
alignment and padding.  This causes problems in lower layers because
drivers which only require memory alignment ends up with adjusted
rq->data_len.  Separate out padding such that padding occurs iff
driver explicitly requests it.

Tomo: restorethe code to update bio in blk_rq_map_user
      introduced by the commit 40b01b9b
      according to padding alignment.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

The meaning of rq->data_len was changed to the length of an allocated
buffer from the true data length. It breaks SG_IO friends and
bsg. This patch restores the meaning of rq->data_len to the true data
length and adds rq->extra_len to store an extended length (due to
drain buffer and padding).

This patch also removes the code to update bio in blk_rq_map_user
introduced by the commit 40b01b9b.
The commit adjusts bio according to memory alignment
(queue_dma_alignment). However, memory alignment is NOT padding
alignment. This adjustment also breaks SG_IO friends and bsg. Padding
alignment needs to be fixed in a proper way (by a separate patch).
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NJens Axboe <axboe@carl.home.kernel.dk>

Interrupts must be disabled if using kmap_atomic(KM_IRQ0), but that was
not the case in a few code paths coming directly from ATA driver
interrupt handlers (which use spin_lock rather than spin_lock_irqsave).
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Back in 2.6.17-rc2, a libata module parameter was added for atapi_dmadir.

That's nice, but most SATA devices which need it will tell us about it
in their IDENTIFY PACKET response, as bit-15 of word-62 of the
returned data (as per ATA7, ATA8 specifications).

So for those which specify it, we should automatically use the DMADIR bit.
Otherwise, disc writing will fail by default on many SATA-ATAPI drives.

This patch adds ATA_DFLAG_DMADIR and make ata_dev_configure() set it
if atapi_dmadir is set or identify data indicates DMADIR is necessary.
atapi_xlat() is converted to check ATA_DFLAG_DMADIR before setting
DMADIR.

Original patch is from Mark Lord.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Mark Lord <mlord@pobox.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Fix libata kernel-doc parameter:

Warning(linux-2.6.25-rc2-git3//drivers/ata/libata-scsi.c:845): No description found for parameter 'rq'
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

This just updates the libata slave configure routine to take advantage
of the block layer drain buffers.  It also adjusts the size lengths in
the atapi code to add the drain buffer to the DMA length so the driver
knows it can rely on it.

I suspect I should also be checking for AHCI as well as ATA_DEV_ATAPI,
but I couldn't see how to do that easily.

tj: * atapi_drain_needed() added such that draining is applied to only
      misc ATAPI commands.
    * q->bounce_gfp used when allocating drain buffer.
    * Now duplicate ATAPI PIO drain logic dropped.
    * ata_dev_printk() used instead of sdev_printk().
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

that provided by the block layer

ATA requires that all DMA transfers begin and end on word boundaries.
Because of this, a large amount of machinery grew up in ide to adjust
scatterlists on this basis.  However, as of 2.5, the block layer has a
dma_alignment variable which ensures both the beginning and length of a
DMA transfer are aligned on the dma_alignment boundary.  Although the
block layer does adjust the beginning of the transfer to ensure this
happens, it doesn't actually adjust the length, it merely makes sure
that space is allocated for transfers beyond the declared length.  The
upshot of this is that scatterlists may be padded to any size between
the actual length and the length adjusted to the dma_alignment safely
knowing that memory is allocated in this region.

Right at the moment, SCSI takes the default dma_aligment which is on a
512 byte boundary.  Note that this aligment only applies to transfers
coming in from user space.  However, since all kernel allocations are
automatically aligned on a minimum of 32 byte boundaries, it is safe to
adjust them in this manner as well.

tj: * Adjusting sg after padding is done in block layer.  Make libata
      set queue alignment correctly for ATAPI devices and drop broken
      sg mangling from ata_sg_setup().
    * Use request->raw_data_len for ATAPI transfer chunk size.
    * Killed qc->raw_nbytes.
    * Separated out killing qc->n_iter.
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Hugh Dickens noticed that SMART commands issued from user space can
end up corupting memory.  The problem occurs if the buffer used to
read data spans two pages.  The reason is that the PIO sector routines
in libata are expecting physically contiguous pages when they do
sector operations, so the left overs on the second page go into the
next physically adjacent page rather than the next page in the sg
mapping.

Fix this by enforcing strict 512 byte alignment on all buffers from
userspace.
Acked-by: NHugh Dickins <hugh@veritas.com>
Acked-by: NJeff Garzik <jeff@garzik.org>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NJeff Garzik <jeff@garzik.org>

Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

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

libata used private sg iterator to handle padding sg.  Now that sg can
be chained, padding can be handled using standard sg ops.  Convert to
chained sg.

* s/qc->__sg/qc->sg/

* s/qc->pad_sgent/qc->extra_sg[]/.  Because chaining consumes one sg
  entry.  There need to be two extra sg entries.  The renaming is also
  for future addition of other extra sg entries.

* Padding setup is moved into ata_sg_setup_extra() which is organized
  in a way that future addition of other extra sg entries is easy.

* qc->orig_n_elem is unused and removed.

* qc->n_elem now contains the number of sg entries that LLDs should
  map.  qc->mapped_n_elem is added to carry the original number of
  mapped sgs for unmapping.

* The last sg of the original sg list is used to chain to extra sg
  list.  The original last sg is pointed to by qc->last_sg and the
  content is stored in qc->saved_last_sg.  It's restored during
  ata_sg_clean().

* All sg walking code has been updated.  Unnecessary assertions and
  checks for conditions the core layer already guarantees are removed.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

ATA_PROT_ATAPI_* are ugly and naming schemes between ATA_PROT_* and
ATA_PROT_ATAPI_* are inconsistent causing confusion.  Rename them to
ATAPI_PROT_* and make them consistent with ATA counterpart.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

atapi_request_sense() is now the only left user of ata_sg_init_one().
Convert it to use sg interface.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Historically word 48 in the identify data was used to mean 32bit I/O
was supported for VLB IDE etc. ATA8 reassigns this word to the Trusted
Computing Group, where it is used for TCG features. This means that
an ATA8 TCG drive is going to trigger 32bit I/O on some systems which
will be funny.

Anyway we need to sort this out ready for ATA8 so:
- Reorder the ata.h header a bit so the ata_version function occurs early
  in it
- Make dword_io check the ATA version
- Add an ATA8 version checking TCG presence test

While we are at it the current drafts have a flaw where it may not be
possible to disable TCG features at boot (and opt out of the trusted
model) as TCG intends because it relies on presence of a different
optional feature (DCS). Handle this in software by refusing the TCG
commands if libata.allow_tpm is not set. (We must make it possible
as some environments such as proprietary VDR devices will doubtless
want to use it to lock up content)

Finally as with CPRM print a warning so that the user knows they may
not be able to full access and use the device.
Signed-off-by: NAlan Cox <alan@redhat.com>

After 9b8e8de7, manage_start_stop configuration depends on valid ATA
device.  Move it into ata_scsi_dev_config().  This was detected by the
coverity checker.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Adrian Bunk <bunk@kernel.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

This patch relaxes the default SCSI DMA alignment from 512 bytes to 4
bytes.  I remember from previous discussions that usb and firewire have
sector size alignment requirements, so I upped their alignments in the
respective slave allocs.

The reason for doing this is so that we don't get such a huge amount of
copy overhead in bio_copy_user() for udev.  (basically all inquiries it
issues can now be directly mapped).
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Prevent libata from starting/stopping non-ATA devices (like ATAPI floppy
drives) as they don't seem to like it:

sd 1:0:1:0: [sdb] Starting disk
ata2.01: configured for PIO2
sd 1:0:1:0: [sdb] Result: hostbyte=0x00 driverbyte=0x08
sd 1:0:1:0: [sdb] Sense Key : 0x2 [current]
sd 1:0:1:0: [sdb] ASC=0x3a ASCQ=0x0
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

None of the drives I have follows what the standard says about
transfer chunk size.  Of the four SATA and six PATA ATAPI devices
tested, four ignore transfer chunk size completely and the ones which
honor it don't behave according to the spec when it's odd.

According to the spec, transfer chunk size can be odd if the amount of
data to transfer equals or is smaller than the chunk size and the
device can indicate the same odd number and transfer the whole thing
at one go with a pad byte appended.  However, in reality, none of the
drives I have does that.  They all indicate and transfer even number
of bytes one byte shorter than the chunk size first; then indicate and
transfer two bytes, which is clearly out of spec.

In addition to unnecessary second PIO data phase, this also creates a
weird problem when combined with SATA controllers which perform PIO
via DMA.  Some of these controllers use actualy number of bytes
received to update DMA pointer so chunks which are sized 4n + 2 makes
DMA pointer off by two bytes.  This causes data corruption and buffer
overruns.

This patch rounds nbytes up to the nearest even number such that ATAPI
devices don't split data transfer for the last odd byte.  This
shouldn't confuse controllers which depend on transfer chunk size as
devices will report the rounded-up number, actually transfer that much
and padding buffer is there to receive them.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Sebastian Kemper reported that issuing CD/DVD commands under libata is
not fully compatible with ide-scsi.  In particular, the
GPCMD_SET_STREAMING was being rejected at the host level in some
instances.

The reason is that libata-scsi insists upon the cmd_len field exactly
matching the SCSI opcode being issued, whereas ide-scsi tolerates
12-byte commands contained within a 16-byte (cmd_len) CDB.

There doesn't seem to be a good reason for us to not be compatible
there, so here is a patch to fix libata-scsi to permit SCSI opcodes so
long as they fit within whatever size CDB is provided.
Signed-off-by: NMark Lord <mlord@pobox.com>
Signed-off-by: NTejun Heo <htejun@gmail.com>

An end to CD-ROM polling (if you have a device that supports AN)...
hooray!
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

ata_scsi_lpm_get() can become static.
Signed-off-by: NAdrian Bunk <bunk@kernel.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Implement ATA_QCFLAG_QUIET which indicates that there's no need to
report if the command fails with AC_ERR_DEV and set it for passthrough
commands.

Combined with previous changes, this now makes device errors for all
direct commands reported directly to the issuer without going through
EH actions and reporting.

Note that EH is still invoked after non-IO device errors to determine
the nature of the error and resume command execution (some controller
requires special care after error to continue).  It just performs
default maintenance after error, examines what's going on, realizes
that it's none of its business and reports the command failure without
logging any error messages.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

ATA_QCFLAG_IO is used to mark commands which are used to perform
regluar IO transfers via block layer.  These commands are assumed to
be valid and taken more seriously during error handling.  Cache flush
is used by regular IO path and necessary for data integrity.  Mark it
with ATA_QCFLAG_IO.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Device Initiated Power Management, which is defined
in SATA 2.5 can be enabled for disks which support it.
This patch enables DIPM when the user sets the link
power management policy to "min_power".

Additionally, libata drivers can define a function
(enable_pm) that will perform hardware specific actions to
enable whatever power management policy the user set up
for Host Initiated Power management (HIPM).
This power management policy will be activated after all
disks have been enumerated and intialized.  Drivers should
also define disable_pm, which will turn off link power
management, but not change link power management policy.

Documentation/scsi/link_power_management_policy.txt has additional
information.
Signed-off-by: NKristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Some commands need post-processing after successful completion.  This
was done in ata_scsi_qc_complete() till now but this has the following
problems.

* Post-command processing gets executed when qc is completed from EH.
  Some qc's are retried from EH with zero err_mask and thus triggers
  unnecessary/incorrect post-command processing.

* Command post processing doesn't belong to SAT layer.

* Link-wide revalidation was scheduled where device revalidation
  suffices.

This patch moves post-command processing to success completion path of
ata_qc_complete() which is travelled iff the command is going to be
completed without passing through EH and updates post-command
processing such that device-specific action is used.  While at it,
restructure code a bit for readability.
Signed-off-by: NTejun Heo <htejun@gmail.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>

Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

LIBATA_MAX_PRD is the maximum number of DMA scatter/gather elements
permitted by the HBA's DMA engine. It's properly set to
q->max_hw_segments via the sg_tablesize parameter.

libata shouldn't call blk_queue_max_phys_segments. Now LIBATA_MAX_PRD
is equal to SCSI_MAX_PHYS_SEGMENTS by default (both is 128), so
everything is fine. But if they are changed, some code (like the scsi
mid layer, sg chaining, etc) might not work properly.

(Addition from Jens) The basic issue is that the physical segment
setting is purely a driver issue. And since SCSI is managing the sglist,
libata has no business changing the setting. All libata should care
about is the hw segment setting.
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Fix libata docbook warnings.

Warning(linux-2.6.23-git8//drivers/ata/libata-scsi.c:3251): No description found for parameter 'dev'
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

After commands which can change device configuration, EH is scheduled
to revalidate and reconfigure the device.  Host link was incorrectly
used unconditionally when scheduling EH action.  This resulted in
bogus revalidation request and mismatched configuration between device
and driver.  Fix it.

This bug was reported by Igor Durdanovic.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Igor Durdanovic <idurdanovic@comcast.net>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Some controller variants snoop the ATAPI length value for Packet
transfers to do state machine and FIFO management. Thus we want to
set it properly, even for cases where it is otherwise meaningless.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

AN serves multiple purposes.  For ATAPI, it's used for media change
notification.  For PMP, for downstream PHY status change notification.
Implement sata_async_notification() which demultiplexes AN.

To avoid unnecessary port events, ATAPI AN is not enabled if PMP is
attached but SNTF is not available.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Kriten Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Some pseudo devices fail PM commands unnecessarily aborting system
suspend.  Implement ATA_HORKAGE_SKIP_PM which makes libata skip PM
commands for these devices.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Controllers which support PMP have various restrictions on which
combinations of commands are allowed to what number of devices
concurrently.  This patch implements ops->qc_defer() which determines
whether a qc can be issued at the moment or should be deferred.

If the function returns ATA_DEFER_LINK, the qc will be deferred until
a qc completes on the link.  If ATA_DEFER_PORT, until a qc completes
on any link.  The defer conditions are advisory and in general
ATA_DEFER_LINK can be considered as lower priority deferring than
ATA_DEFER_PORT.

ops->qc_defer() replaces fixed ata_scmd_need_defer().  For standard
NCQ/non-NCQ exclusion, ata_std_qc_defer() is implemented.  ahci and
sata_sil24 are converted to use ata_std_qc_defer().

ops->qc_defer() is heavier than the original mechanism because full qc
is prepped before determining to defer it, but various information is
needed to determine defer conditinos and fully translating a qc is the
only way to supply such information in generic manner.

IMHO, this shouldn't cause any noticeable performance issues as

* for most cases deferring occurs rarely (except for NCQ-aware
  cmd-switching PMP)
* translation itself isn't that expensive
* once deferred the command won't be repeated until another command
  completes which usually is a very long time cpu-wise.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Update AN support in preparation of PMP support.

* s/ata_id_has_AN/ata_id_has_atapi_AN/
* add AN enabled reporting during configuration
* add err_mask to AN configuration failure reporting
* update LOCKING comment for ata_scsi_media_change_notify()
* check whether ATA dev is attached to SCSI dev ata_scsi_media_change_notify()
* set ATA_FLAG_AN in ahci and sata_sil24
Signed-off-by: NTejun Heo <htejun@gmail.com>
Cc: Kriten Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>