Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Fix host reset escalation killing all IOs.

SLI-3 adapters will use a new host template. The template differs
from SLI-4 adapters in that it does not have an eh_host_reset_handler.

Lpfc has traditionally never had a host_reset. The host reset
handler was added when we ran into a stuck hardware condition on a
SLI-4 adapter. The host_reset will reset and reinit the pci function,
clearing the hardware condition.

Unfortunately, the host reset handler uses attach/detach code paths,
which makes scsi_add_host() and scsi_remove_host() calls. Meaning, a
host_reset will completely remove the scsi_host from the system. As a
new call to scsi_add_host() is made, the shost# changes, which results
in completely new scsi_devices and device names. All the older scsi
devices on the old shost# are now orphaned and unrecoverable.

We realize we need to re-implement the host_reset_handler so the scsi_host
stays registered across the host_reset, but that will be a rather
lengthy effort. In the short term, we had an immediate need to restore
the SLI-3 devices to their working behavior, with the easiest path being
to remove their host_reset handler.
Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

This will detect and send an async event if overtemp is detected
Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NDick Kennedy <dick.kennedy@emulex.com>
Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NJames Smart <james.smart@emulex.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NEwan D. Milne <emilne@redhat.com>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

The loop in aha1542_outb with double-check is no longer needed, remove it.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Bus reset always fails because aha1542_reset waits for the controller to assert
the INIT bit (Mailbox Initialization Required) which it never does. This bit is
asserted only after host reset.

Remove the requirement for INIT bit (we really need only the IDLE bit).
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Remove aha1542_lock and use host_lock instead.
Remove interrupt and queuecommand function wrappers.
Remove locking from lowlevel _out and _in functions, they now can
onle be called (at runtime) with host_lock being held.
Remove ssleep(4) in aha1542_reset as we can't sleep while holding a spinlock.
It's useless anyway as wait_mask will wait until the controller is idle and
kernel waits for 10 seconds (HOST_RESET_SETTLE_TIME) after that.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Enabling DEBUG disables write commands and devices with ID > 1.
Remove this "feature" to allow real debugging.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Remove DEB macro and join ifdef DEBUG blocks
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Pass struct Scsi_Host *sh and dma parameters instead of index to
aha1542_set_bus_times.
This is required for printk conversion.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Pass struct Scsi_Host * to functions instead of base address.
This reduces the number of parameters and is also required for printk
conversion.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Fix include guard in header file and remove useless changelog
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Make Scsi_Host variable names consistent - use sh everywhere.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Use struct scsi_cmnd instead of Scsi_Cmnd and also rename the variables
to get rid of SC prefix
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Make sure that there's no variable named cmd
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Simplify aha1542_biosparam, use sector_div, remove unused BIOS_TRANSLATION_1632.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Remove __setup and introduce separate io, bus_on, bus_off and dma_speed
module parameters.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Cleanup hw_init, use goto for error handling.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

aha1542_out call is always followed by wait_mask.
Move the call into aha1542_out to simplify code.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Split out bus times related block of aha1542_hw_init into separate function
aha1542_set_bus_times.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

aha1542_host_reset and aha1542_bus_reset are almost same, merge them
into aha1542_reset
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

aha1542_restart does nothing (except useless printk), remove it
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Remove gotos that are no longer needed.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

aha1542_out are in fact two separate functions.
Split them into aha1542_out and aha1542_outb to simplify the code.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Unify aha1542_in and aha1542_in1 functions, they differ only in timeout
and printk.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Convert the ugly WAIT and WAITd macros into wait_mask function.
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>

Convert aha1542_intr_reset macro to inline function
Signed-off-by: NOndrej Zary <linux@rainbow-software.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJames Bottomley <JBottomley@Odin.com>