The comedi core module optionally allocates some legacy board minor
devices on module load and cleans these up on module exit.  These are
used for manual configuration of comedi boards (for those low-level
comedi drivers that support manual configuration - mainly for ISA
boards).  Other board minor devices are created and destroyed
dynamically in response to bus device probe and remove requests.  The
ioctl used for manual configuration (attachment) and removal
(detachment) of devices is COMEDI_DEVCONFIG, but that works for any
board minor device, including those that were originally created
dynamically.

If the COMEDI_DEVCONFIG ioctl is used to manually detach an
automatically created and attached device, commit
7d3135af ("staging: comedi: prevent
auto-unconfig of manually configured devices") ensures that the board
minor will no longer be automatically detached and destroyed by a bus
device remove request.  From that point on the board minor behaves more
like one of the comedi "legacy" board minors.  (There would be some
justification for destroying the board minor instead, but I'd rather
leave that decision until removal of board minors has been
made safer than it currently is.)  Although the board minor behaves more
like a legacy board minor, it is not currently cleaned up on module
exit.  In fact, the module exit code will bug out because this board
minor has not been cleaned up.

Change comedi_cleanup_legacy_minors() (called from the module exit code,
and from the module init code on error) to clean up all board minors.
Rename the function to comedi_cleanup_board_minors() to reflect the
change in functionality.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Comedi has two sorts of minor devices:
(a) normal board minor devices in the range 0 to
COMEDI_NUM_BOARD_MINORS-1 inclusive; and
(b) special subdevice minor devices in the range COMEDI_NUM_BOARD_MINORS
upwards that are used to open the same underlying comedi device as the
normal board minor devices, but with non-default read and write
subdevices for asynchronous commands.

The special subdevice minor devices get created when a board supporting
asynchronous commands is attached to a normal board minor device, and
destroyed when the board is detached from the normal board minor device.
One way to attach or detach a board is by using the COMEDI_DEVCONFIG
ioctl.  This should only be used on normal board minors as the special
subdevice minors are too ephemeral.  In particular, the change
introduced in commit 7d3135af ("staging:
comedi: prevent auto-unconfig of manually configured devices") breaks
horribly for special subdevice minor devices.

Since there's no legitimate use for the COMEDI_DEVCONFIG ioctl on a
special subdevice minor device node, disallow it and return -ENOTTY.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Reported-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

FWSERIAL_TTY_START_MINOR was removed. The minor_start is allocated
by tty_alloc_driver().
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Use tty_alloc_driver() instead.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Add the following file hierarchy to debugfs:

  <debugfs>-+
            +- firewire_serial -+- <unit> -+- peers
                                |          +- stats
                                |
                                +- <unit> -+- peers
                                           +- stats

The 'peers' file (read-only) contains status and configuration
info for attached peers for the given fwserial unit.

The 'stats' file (read-only) contains statistics and data profiling
information for each tty port for the given fwserial unit.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Factor out extra stats, data profiles, debugging info and peer info
from procfs file in preparation for using debugfs instead.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Register a second tty driver to create loopback devices for
each firewire node. Note that the loopback devices are numbered
from 0; the tty->index is transformed when used to index the
port table.

Remove the hack that previously enabled this.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

After sending the unplug response, release the port even if an
error occurred.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

When a port has been reserved in an attempt to connect to a peer
but that attempt does not succeed, releasing the port should not
reset the port line status. Although resetting is functionally
harmless, it can appear as if a remote peer dropped carrier to a
port it was not attached to (which can be confusing).
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The firewire core does not require or want the suggested helper fns;
drop suggestion from TODO file.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Devices which are OHCI v1.0/ v1.1/ v1.2-draft compliant or
RFC 2734 compliant are required by specification to support
max_rec of 8 (512 bytes) or more. Accept reported value.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Since peer->max_payload is now limited to 1394-2008 spec maximum
of 4096, the port->max_payload limit can now be simplified.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Self-limiting asynchronous bandwidth (via reducing the payload)
is not necessary and does not work, because
 1) asynchronous traffic will absorb all available bandwidth (less that
    being used for isochronous traffic)
 2) isochronous arbitration always wins.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The patch ae926051: "staging: ozwpan: Added USB HCD
implementation" from Feb 20, 2012, leads to the following warning:
drivers/staging/ozwpan/ozhcd.c:1094 oz_hcd_heartbeat()
	 warn: what is this condition about? 'ep->buffered_units * 50'
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Rupesh Gujare <rupesh.gujare@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Replace the various command strings by named constants.
Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The EC command for enable/disable is not an EC command. Instead it needs
to be send to the mouse.
Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Cleanup toggle of global event reporting by moving it to its own function.
This simplifies the following cleanup.
Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Recently, Matt Sealey reported he fail to build zsmalloc caused by
using of local_flush_tlb_kernel_range which are architecture dependent
function so !CONFIG_SMP in ARM couldn't implement it so it ends up
build error following as.

  MODPOST 216 modules
  LZMA    arch/arm/boot/compressed/piggy.lzma
  AS      arch/arm/boot/compressed/lib1funcs.o
ERROR: "v7wbi_flush_kern_tlb_range"
[drivers/staging/zsmalloc/zsmalloc.ko] undefined!
make[1]: *** [__modpost] Error 1
make: *** [modules] Error 2
make: *** Waiting for unfinished jobs....

The reason we used that function is copy method by [1]
was really slow in ARM but at that time.

More severe problem is ARM can prefetch speculatively on other CPUs
so under us, other TLBs can have an entry only if we do flush local
CPU. Russell King pointed that. Thanks!
We don't have many choices except using flush_tlb_kernel_range.

My experiment in ARMv7 processor 4 core didn't make any difference with
zsmapbench[2] between local_flush_tlb_kernel_range and flush_tlb_kernel_range
but still page-table based is much better than copy-based.

* bigger is better.

1. local_flush_tlb_kernel_range: 3918795 mappings
2. flush_tlb_kernel_range : 3989538 mappings
3. copy-based: 635158 mappings

This patch replace local_flush_tlb_kernel_range with
flush_tlb_kernel_range which are avaialbe in all architectures
because we already have used it in vmalloc allocator which are
generic one so build problem should go away and performane loss
shoud be void.

[1] f553646a, zsmalloc: add page table mapping method
[2] https://github.com/spartacus06/zsmapbench

Cc: stable@vger.kernel.org
Cc: Dan Magenheimer <dan.magenheimer@oracle.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Konrad Rzeszutek Wilk <konrad@darnok.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Seth Jennings <sjenning@linux.vnet.ibm.com>
Reported-by: Matt Sealey <matt@genesi-usa.com>
Signed-off-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Found with coccicheck.
The semantic patch that makes this change is available
in scripts/coccinelle/api/memdup.cocci.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Rob Clark <rob@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This fixes a checkpatch.pl issue of
'braces {} are not necessary for single statement blocks'
Signed-off-by: Ryo Munakata <ryomnktml@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Found with coccicheck.
The semantic patch that makes this change is available
in scripts/coccinelle/api/memdup.cocci.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Right now ZS_SIZE_CLASS_DELTA is hardcoded to be 16.  This
creates 254 classes for systems with 4k pages. However, on
PPC64 with 64k pages, it creates 4095 classes which is far
too many.

This patch makes ZS_SIZE_CLASS_DELTA relative to PAGE_SIZE
so that regardless of the page size, there will be the same
number of classes.
Acked-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Seth Jennings <sjenning@linux.vnet.ibm.com>
Acked-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Assigning a string is really bad,
and since we only have 1 char strings here we
can simply use a char to store the value and change the format specifier.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Since `get_device()` and `put_device()` can take a NULL device
parameter, `comedi_set_hw_dev()` can be simplified to always call
`get_device()` for the new, possibly NULL hardware device, and
`put_device()` for the old, possibly NULL hardware device.  As long as
we do it in that order, there shouldn't be any problem with
`kref_release()` getting called unexpectedly when the new hardware
device is the old hardware device.

Simplify `comedi_set_hw_dev()` and update the comment because the
function is used for additional purposes since the old comment was
written.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

For comedi devices that support asynchronous commands on some of their
subdevices, the comedi core creates extra device files for each of those
subdevices of the form "/dev/comedi%i_subd%i".  These use the same
comedi device as the corresponding "/dev/comedi%i" but override the
default "read" and "write" subdevice for the device.  Currently it
overrides both the read and write subdevice, but it only makes sense to
override the "read" subdevice if the subdevice supports "read" commands,
and to override the "write" subdevice if the subdevice supports "write"
commands.

In `comedi_alloc_subdevice_minor()`, only set `info->read_subdevice`
non-NULL if the subdevice has the `SDF_CMD_READ` flag set, and only set
`info->write_subdevice` non-NULL if the subdevice has the
`SDF_CMD_WRITE` flag set.  (`comedi_read_subdevice(info)` will use the
device's default read subdevice if `info->read_subdevice` is NULL.
`comedi_write_subdevice(info)` will use the device's default write
subdevice if `info->write_subdevice` is NULL.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The LABPC_DEBUG define is not used in any of the code. Just
remove the #undef.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

For aesthetic reasons, reformat the boardinfo declaration and
add some whitespace. Remove all the information that is set to
'0' as this is the default.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'thisboard' macro relies on a local variable having a specific
name and yields a pointer derived from that local variable.

Replace the macro with a local variable and use the comedi_board()
helper to get the pointer.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Move the comedi_driver declaration down in the file. This removes
the need for the forward declaration.

For aesthetic reasons, add some whitespace to the declaration and
remove the unnecessary '&' before the function names. They are
already addresses.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The private data, struct local_info_t, is not being used in the
driver. Remove it as well as the kzalloc/kfree.

Also, don't set the 'pcmcia_cur_dev' variable unless the pcmcia
configuration is successful.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Absorb the code from labpc_config() into this function and properly
return the error if the configuration fails.

Also, remove the dev_dbg() function trace message.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

For aesthetic reasons, move all the pcmcia_driver functions so they
are near the pcmcia_driver declaration. This also removes the need
for some of the forward declarations.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The pcmcia_driver suspend and resume functions in this driver
don't do anything. Since they are optional just remove them.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The pcmcia_driver suspend and remove functions set the 'stop' variable
and the resume function clears it. Nothing in the comedi_driver code
uses the 'stop' variable.

Just remove it so we can get rid of the suspend/resume.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The function simply calls pcmcia_disable_device(). Remove it and
just call pcmcia_disable_device() where needed.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>