Implement all states in scic_remote_device_start() and delete the state
handler.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

While reducing indentation commits 7ab92c9e "isci: make a
remote_node_context a proper member of a remote_device", 0879e6a6 "isci:
merge remote_device substates into a single state machine" broke
handling of situations where i/o's successfully started at the port
level need to terminated when the remote_node declines to start the i/o.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

A function call to dereference a pointer is a tad much.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

A substate is just a state, so uplevel the smp and stp device substates.
Three tricks at work here:

1/ scic_sds_remote_device_ready_state_enter: needs to know the the device type
   so it can immediately transition to a stp or smp ready substate.

2/ scic_sds_remote_device_ready_state_exit: needs to know the device type. In
   the ssp case the device is no longer ready, in the stp, and smp case we have
   simply exited to a ready "substate".

3/ scic_sds_remote_device_resume_complete_handler: The one location
   where we directly check the current state against
   SCI_BASE_REMOTE_DEVICE_STATE_READY needed to comprehend the possible ready
   substates.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The sci_object.h file was removed. No sci_base_object
is now in the code.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_request and was replaced by a pointer to
struct isci_request.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_remote_node_context.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_remote_device.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
[cleaned up sci_dev_to_idev]
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_port and was replaced by a pointer to
struct isci_port.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_phy and was replaced by a pointer to
struct isci_phy.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The 'struct sci_base_object' was removed from the struct
scic_sds_controller and was replaced by a pointer to
struct isci_host.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Changed any occurrence of struct sci_base_object into void.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Updated SCU AFE initialization values accordingly to the recipe 10.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

We are logging excessive output when hot unplug from expander. Moving
that to debug.
Signed-off-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

If the scu efi driver is disabled but the option-rom is enabled (during an efi
boot) allow the code to fallback to scanning legacy option-rom space for the
parameters.
Reported-by: NYinghai Lu <yinghai.lu@oracle.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Removing not used / bit-rotten ATAPI code. This needs to go back
and debugged at a later date.
Signed-off-by: NDave Jiang <dave.jiang@intel.com>
[reflow against devel, delete dead sati headers]
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Longer to type than the open-coded equivalent.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The sas address can be retrieved from the domain device and then
converted to the always little-endian format in the remote node context.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

An lldd need never look at the contents of an smp_discover_response frame.
Kill the remaining locations where isci is looking at it:

1/ covering for expanders that do not set the stp_attached bit (already
   handled by sas_ex_discover_end_dev)
2/ an overkill method to notifiy the rest of the driver about remote_device
   sas addresses
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

This is step 1 of removing the contortions to:
1/ unparse expander phy data into a smp discover frame
2/ open-code-parse the smp discover fram into a domain_device.dev_type equivalent

libsas has already spent cycles determining the dev_type, so now that
scic_sds_remote_device is unified with isci_remote_device we can
directly reference dev_type.

This might also change multi-level expander detection as we previously only
looked at dev_type == EDGE_DEV and we did not consider the FANOUT_DEV case.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The construction routines scic_remote_device_[de]a_construct both reference
the need to call scic_remote_device_construct first.  Delete that comment and
just have them call it explicitly, also:
* move the comments from header to source
* delete dead references to scic_[de]a_remote_device_add_phy
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Now that the core/lldd remote_device data structures are nominally unified
merge the corresponding sources into the top-level directory.  Also move the
remote_node_context infrastructure which has no analog at the lldd level.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Make it explicit that isci_remote_device and scic_sds_remote_device are
one in the same object.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Now that they are one in the same object remove the back pointer reference
in favor of container_of.
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

A rnc object has the same lifetime as its associated remote_device.  It might
get re-initialized, but a remote device always has an rnc member.  Preparation
for unifying scic_sds_remote_device and isci_remote_device
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Remove the extra logic to poll each controller for interrupts, that's
the core's job for shared interrupts.

While testing noticed that a number of interrupts fire while waiting for
the completion tasklet to run, so added an irq-ack.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Removed any instances of the_* and this_* to variable names that are more
meaningful and tell us what they actually are.
Signed-off-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

If the platform specifies invalid parameters warn the user and fallback to
internal defaults rather than fail the driver load altogether.
Reported-by: NYinghai Lu <yinghai.lu@oracle.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

The element_length is 2 bytes.
Reported-by: NYinghai Lu <yinghai.lu@oracle.com>
Acked-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Removes unnecessary usage of BUG_ON macro, excluding core directory.
In some cases macro is unnecesary, check is done in caller function.
In other cases macro is replaced by if construction with
appropriate warning.
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
[changed some survivable bug conditions to WARN_ONCE]
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Clean warnings and errors reported by sparse tool.

request.c:430:50: warning: mixing different enum types
remote_device.c:534:39: warning: symbol 'flags' shadows an earlier one
task.c:495:44: warning: mixing different enum types
scic_sds_controller.c:2155:24: warning: mixing different enum types
scic_sds_controller.c:2272:36: warning: mixing different enum types
scic_sds_controller.c:2911:38: warning: incorrect type in initializer (different address spaces)
scic_sds_controller.c:2913:25: warning: incorrect type in argument 2 (different address spaces)
scic_sds_request.c:875:34: warning: cast removes address space of expression
scic_sds_request.c:876:123: warning: incorrect type in argument 2 (different address spaces)
scic_sds_port.c:585:51: warning: incorrect type in assignment (different address spaces)
scic_sds_port.c:712:9: warning: incorrect type in argument 2 (different address spaces)
scic_sds_port.c:1770:25: warning: incorrect type in argument 2 (different address spaces)
Signed-off-by: NBartosz Barcinski <Bartosz.Barcinski@intel.com>
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
[fixed up some false positives and misconversions]
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Drop duplicated enum definition.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Merge struct sci_base_phy into scic_sds_phy.  Until now sci_base_phy was
referenced using scic_sds_phy->parent field.

'sci_base_phy' state machine handlers were also merged into scic_sds_phy
state handlers.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMaciej Trela <Maciej.Trela@intel.com>
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Merge struct sci_base_port into scic_sds_port.  Until now sci_base_port
was referenced indirectly with scic_sds_port->parent field.

'sci_base_port' state machine handlers were also incorporated into
scic_sds_port handlers.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMaciej Trela <Maciej.Trela@intel.com>
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Merge struct sci_base_remote_device into scic_sds_remote_device.  As for
now sci_base_remote_device was accessed indirectly using
scic_sds_remote_device->parent field.  Both machine state handlers are
also merged together.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMaciej Trela <Maciej.Trela@intel.com>
Signed-off-by: NMaciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Remove the state handler indirections for the scic_controller, and replace
them with procedural calls that check for the correct state first.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Remove the insane infrastructure for preallocating coheren DMA regions,
and just allocate the memory where needed.  This also gets rid of the
aligment adjustments given that Documentation/DMA-API-HOWTO.txt sais:

  "The cpu return address and the DMA bus master address are both
   guaranteed to be aligned to the smallest PAGE_SIZE order which
   is greater than or equal to the requested size.  This invariant
   exists (for example) to guarantee that if you allocate a chunk
   which is smaller than or equal to 64 kilobytes, the extent of the
   buffer you receive will not cross a 64K boundary."
Signed-off-by: NChristoph Hellwig <hch@lst.de>
[djbw: moved allocation from start to init, re-add memset]
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Instead of filling up tables with default handlers call the default
handler in the only caller.

IMHO the whole state handlers concept is not very suitable for the
isci request.  For example there is a single real instance of the
start handler, and we'd be much better off just having a check for
the right state in the only caller, than all this mess.  It's
quite similar for the abort handler as well.

Even the actual state machine has a lot of states that are rather
pointless.  The initial and constructed states are not needed at all
as the request is not reachable for calls before it's fully set up and
started.  And the abort state should be replaced with an abort actions
and a state transition to the completed state.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Merge struct sci_base_request into scic_sds_request, and also factor the two
types of state machine handlers into one function.  While we're at it also
remove lots of duplicate incorrect kerneldoc comments for the state machine
handlers.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>