The SCSI-midlayer's fast-fail codes consider an DID_ERROR status
as a driver-error and the failed I/O would then be retried in the
midlayer without being fast-failed to dm-multipath.  DID_BUS_BUSY
status returns would induce unneeded path-failures events being
propagated to the DM/MD.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Changing a lun's queue depth (/sys/block/sdX/device/queue_depth)
isn't sticky when the device is connected via a QLogic fibre
channel adapter.

The QLogic qla2xxx fibre channel driver dynamically adjusts a
lun's queue depth.  If a user has a specific need to limit the
number of commands issued to a lun (say a tape drive, or a shared
raid where the total commands issued to all luns is limited at
the controller level, for example) and writes a limiting value to
/sys/block/sdXX/device/queue_depth, the qla2xxx driver will
silently and gradually increase the queue depth back to the
driver limit of ql2xmaxqdepth.  While reducing this value (module
parameter) or increasing the interval between ramp ups
(ql2xqfullrampup) offers the potential for a work around it would
be better to have the option of just disabling the dynamic
adjustment of queue depth.

This patch implements an "off switch" as a module parameter.
Signed-off-by: NMichael Reed <mdr@sgi.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Set the module parameter ql2xmultique_tag to 1 to enable this
feature. In this mode, the total number of response queues
created is equal to the number of online cpus. Turning the block
layer's rq_affinity mode on enables requests to be routed to the
proper cpu and at the same time it enables completion of the IO
in a response queue that is affined to the cpu in the request
path.
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Set the number of request queues to the module paramater
ql2xmaxqueues.  Each vport gets a request queue. The QoS value
set to the request queues determines priority control for queued
IOs. If QoS value is not specified, the vports use the default
queue 0.
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

When IRQs are shared by multiple controllers and if the first one
to register does not disable the IRQ, then IRQ will be enabled
for all other controllers by default, irrespective of their
setting. With IRQF_DISABLED registration, the driver interrupt
routine was called with interrupt enabled always. Disbaling the
registration with IRQF_DISABLED, since driver code is re-entrant
safe and all critical sections are guarded with interrupt safe
locks.
Signed-off-by: NGiridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

General cleanup of extraneous/legacy crud.

Additional cleanups and
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Original code used an overabundance of indirect pointers to
function helpers.  Instead, the driver can exploit the immutable
properties of a queue's ISP-association and ID, which are both
known at queue initialization-time.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

To ensure smooth operations amongst the FCoE and NIC side
components of the ISP81xx chip, the FCoE driver (qla2xxx) must
ensure the 10gb NIC driver (qlge) does not timeout waiting for
IDC (Inter-Driver Communication) acknowledgments.  The
acknowledgment requirements are trivial -- a simple mirroring of
incoming mailbox registers during the AEN to a process-context
capable mailbox command.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Correct response-queue-0 processing by instructing the firmware
to run with interrupt-handshaking disabled, similarly to what is
now done for all non-0 response queues.  Since all
response-queues now run in the same mode, the driver no longer
needs the hot-path 'is-disabled-HCCR' test.
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Codes to support new FCoE boards.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Original code would inadvertanly place I/Os on the default
request-queue.  Also, correctly pass in the proper MSI-X vector
during response-queue initialization.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Software should not touch this region of flash, as the firmware
will be the only writer and consumer of the region.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Following changes have been made:
 1. Scan outstanding commands only in the queue where it is submitted
 2. Update queue registers directly in the fast path
 3. Queue specific BAR is remapped only for multiq capable adapters
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Following changes have been made.
1. qla_hw_data structure holds an array for request queue pointers,
and an array for response queue pointers.
2. The base request and response queues are created by default.
3. Additional request and response queues are created at the time of vport
creation. If queue resources are exhausted during vport creation, newly
created vports use the default queue.
4. Requests are sent to the request queue that the vport was assigned
in the beginning.
5. Responses are completed on the response queue with which the request queue
is associated with.

[fixup memcpy argument reversal spotted by davej@redhat.com]
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Following changes have been made:

1. Outstanding commands are based on a request queue, scsi_qla_host
   does not maintain it anymore.

2. start_scsi is accessed via isp_ops struct instead of direct
   invocation.

3. Interrupt registrations are done using response queue instead of
   device id.
Signed-off-by: NAnirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

This has qla2xxx use the new transport error values instead of
DID_BUS_BUSY. I am not sure if all the errors
in qla_isr.c I changed are transport related. We end up blocking/deleting
the rport for all of them so it is better to use the new transport error since
the fc classs will decide when to fail the IO.

With this patch if I pull a cable then IO that had reached
the driver, will be failed with DID_TRANSPORT_DISRUPTED (not including
tape). The fc class will then fail the IO when the fast io fail tmo
has fired, and the driver will flush any other commands running.
Signed-off-by: NMike Christie <michaelc@cs.wisc.edu>
Cc: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

As the driver is only interested in bits 0-9 of the 1st
RSCN-payload word:

	rsvd[15:14]RscnEventQualifier[13:10]Fmt[9:8]Domain[7:0]
	Area[15:8]Alpa[7:0]
Signed-off-by: NRavi Anand <ravi.anand@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Add additional tightening of residual-count handling (originally
from commit 6acf8190) where the
driver should discard any lower SCSI-status during
firmware/transport residual-count mismatches.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Fix "occured" spelling errors.  Most of these are in comments, which
I wouldn't normally bother with, but a couple are in printks, which
irritate me more.  So I just fixed them all at the same time.
Signed-off-by: NBjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Josip Rodin noted
(http://article.gmane.org/gmane.linux.ports.sparc/10152) the
driver oopsing during registration of an rport to the
FC-transport layer with a backtrace indicating a dereferencing of
an shost->shost_data equal to NULL.  David Miller identified a
small window in driver logic where this could happen:

    > Look at how the driver registers the IRQ handler before the host has
    > been registered with the SCSI layer.
    >
    > That leads to a window of time where the shost hasn't been setup
    > fully, yet ISRs can come in and trigger DPC thread events, such as
    > loop resyncs, which expect the transport area to be setup.
    >
    > But it won't be setup, because scsi_add_host() hasn't finished yet.
    >
    > Note that in Josip's crash log, we don't even see the
    >
    >         qla_printk(KERN_INFO, ha, "\n"
    >             " QLogic Fibre Channel HBA Driver: %s\n"
    >             "  QLogic %s - %s\n"
    >             "  ISP%04X: %s @ %s hdma%c, host#=%ld, fw=%s\n",
    >  ...
    >
    > message yet.
    >
    > Which means that the crash occurs between qla2x00_request_irqs()
    > and printing that message.

Close this window by enabling RISC interrupts after the host has
been registered with the SCSI midlayer.
Reported-by: NJosip Rodin <joy@entuzijast.net>
Cc: Stable Tree <stable@kernel.org>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

The executing-HA of an SRB can be referenced from the sp->fcport.
Use this correct value while processing status-continuation data
and abort processing.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

* Use correct 'ha' to mark a device lost from ISR.
  I/Os will always be returned on the physical-HA.
  qla2x00_mark_device_lost() should be called with the HA bound
  to the fcport.
* Mark *all* devices lost during ISP-ABORT (bighammer).

These fixes correct issues discovered locally where during
link-perturbation and heavy vport-I/O fcport/rport states would
stray and an rport's scsi-target lost (timed-out).
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

By allowing the qla2x00_alert_all_vps() to manage per-vport
recognition of the MBA.
Signed-off-by: NSeokmann Ju <seokmann.ju@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

This reverts commit bd2a1846.
The original (prior to the reverted commit) code was correct.
Additionally, the vp_idx should be checked during MBA_PORT_UPDATE
in order for proper handling to take place for a given vport.
Signed-off-by: NSeokmann Ju <seokmann.ju@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

This reverts commit 8084fe16.
The midlayer should be given the oppotunity to interpret the
check-condition and based on scsi_cmnd->resid determine if a
transfer should be retried or failed.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Both MSI-X vector handlers attempt to acquire the HA's
hardware_lock.  This though requires that interrupts be
disabled/enabled during acquisition and release of the spinlock.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Original code would (incorrectly) only re-register after a
loop-down condition.  Also, FDMI registration should be enabled
by default.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Changes are added to the driver so that it can behave properly
upon having asynchronous events, for example, addition of target
devices to the VPs.
Signed-off-by: NSeokmann Ju <seokmann.ju@qlogic.com>
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NRavi Anand <ravi.anand@qlogic.com>
Additional cleanups and
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Strip unused (DEBUG-ONLY) enabled functions, inlines, useless
wrappers, and unused DPC flags from the code.  Another step in
the migration towards a cleaner (less-crusty) driver.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Recent ISPs have a region within FLASH which acts as a repository
for the logging of serious hardware and software failures.
Currently, the region is large enough to support up to 255
entries.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Supported events include LIP, LIP reset, RSCN, link up, and link
down.

To support AEN (and additional forthcoming features), we also
introduce a simple deferred-work construct to manage events which
require a non-atomic sleeping-capable context.  This work-list is
processed as part of the driver's standard DPC routine.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

There's no need to use the heavier (albiet safer)
*_irq[save|restore]() locking primitives within the driver's
interrupt handlers, interrupts are guaranteed to be
non-reentrant.  Use lightweight spin_lock() and spin_unlock()
primitives while acquiring the hardware_lock.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

For recent ISPs, software must detect OVERRUN conditions by
checking the SS_RESIDUAL_OVER bit during CS_COMPLETE handling.
Update the driver to perform this check, which is consistent with
what earlier firmwares did by explicitly cracking open the
FCP_RSP statuses and returning an CS_DATA_OVERRUN.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

There's no functional change involved with this update, instead
it simply migrates the "set cleared interrupt state" codes to a
more approprate method, qla2x00_request_irqs(), and cleans-up the
driver's probe() logic.
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Instead of abusing the semaphore interfaces for mailbox command
completions.

Additional cleanups and
Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Signed-off-by: NAndrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>