First introduced in kernel 2.3.99-pre3 (added to ide_dmaproc) and never used.

Introduced in kernel 2.5.63 and never used.

Introduced in kernel 2.5.35 and never used.

Fix CONFIG_HPT34X_AUTODMA to depend directly on CONFIG_EXPERIMENTAL
and finally remove CONFIG_IDEDMA_PCI_WIP.

Leftover after switch to IDE DMA Model-Firmware blacklist in kernel 2.5.63.

Some devices need more time to become non-busy after command completion,
so use the same timeout in wait_drive_not_busy() (called from task_in_intr()
if CONFIG_IDE_TASKFILE_IO=y) as in drive_cmd_intr().

Thanks to Mark Haverkamp <markh@osdl.org> for help in fixing this.

into home.osdl.org:/home/torvalds/v2.5/linux

into nuts.davemloft.net:/disk1/BK/net-2.6

into home.osdl.org:/home/torvalds/v2.5/linux

	* don't take the unnecessary ->host_lock around ->scsi_done() call.

	* switched to proper parport_driver.

	* switched to use ->hostdata to get ppa_struct from Scsi_Host, end
of messing with ppa_hosts[]

	* fixed missing wakeups in ppa_pb_claim()/ppa_wakeup() - if the
former had been called just as current holder of port was giving it up,
we could set "I'm waiting" flag too late.  Cleaned up the timeout logics.

	* starting to kill ppa_detect - we move the "probe a single port"
logics into a separate function and shift scanning into ppa_driver_init().
Later that will give us a parport_driver ->attach().

	* killed include of scsi_module.h and switched from scsi_register()
to scsi_alloc_host().  We still keep the old detect logics at that point,
it will be gone later.

	* killed ppa_pb_claim() call in ppa_init()

	* switched ppa to passing around references to ppa_struct instead
of host numbers.

	* ppa.c fed through Lindent, functions unused elsewhere (by now -
	  all of them) made static.

	* switched to proper parport_driver.

	* switched to use ->hostdata to get imm_struct from Scsi_Host, end
of messing with imm_hosts[]

	* fixed missing wakeups in imm_pb_claim()/imm_wakeup() - if the
former had been called just as current holder of port was giving it up,
we could set "I'm waiting" flag too late.  Cleaned up the timeout logics.

	* starting to kill imm_detect - we move the "probe a single port"
logics into a separate function and shift scanning into imm_driver_init().
Later that will give us a parport_driver ->attach().

	* killed include of scsi_module.h and switched from scsi_register()
to scsi_alloc_host().  We still keep the old detect logics at that point,
it will be gone later.

	* killed imm_pb_claim() call in imm_init()

	* switched imm to passing around references to imm_struct instead
of host numbers.

	* imm.c fed through Lindent, functions unused elsewhere (by now -
	  all of them) made static.

paride has a ->private field for use by low-level drivers.  It's not
64bit-clean - it's int, but is used to store a pointer (bpck6).
Fix: replace with unsigned long, update typecasts in users.

pd_identify() converted to pd_special_command().

pd_init_units() merged into pd_detect(), pd_drive_count made local to
pd_detect().

Now disk->present is not needed - we can always check for disk->gd.
Field removed, pd_probe_drive() became void(struct pd_unit *), pd_detect()
cleaned up,

Moved allocation and setup (but not adding) of gendisk into
pd_probe_drive().

Switched everything except pd_identify() to use of special requests.
New helper: pd_special_command() - builds special request, feeds it
into queue and waits for completion.
Moved pi_connect()/pi_disconnect() to the beginning/end of pi_identify.

Reordered stuff
made pd_ready() inlined
renamed ps_timeout to pd_timeout
renamed ps_tq to fsm_tq
renamed ps_set_intr to fsm_schedule
removed unused pd_wait_open.

Moved calls of ->connect and ->disconnect into run_fsm().  The only
change is that now do_pd_start_io() is ran when we are connected.

Equivalent transformation:
	now that all phase functions are called claimed, we can
	simplify the logics around claiming - new variable (pd_claimed)
	keeps track of claim/unclaim we'd done.  We check it in the
	beginning of the loop and do claiming if it's not set.

Equivalent transformation:
	* phase = do_pd_io replaced with phase = NULL, corresponding
	  check + open-coded equivalent added in run_fsm loop.
	* do_pd_io() is gone.

Equivalent transformation:
	next_request() call expanded.

Equivalent transformation:
	call of pi_do_claimed() in run_fsm() expanded; recursive
	call of run_fsm() replaced with loop.

Equivalent transformation:
	pi_unclaim() moved from phase functions to run_fsm().

Equivalent transformation:
	* calls of pi_disconnect() in phase functions expanded.
	* since both do_pd_read_start() and do_pd_write_start()
	  are always called claimed (as the matter of fact,
	  all phase functions except do_pd_io() are), we can
	  replace call of pi_connect() with direct method call -
	  pi_claim() is no-op there.