For aesthetics, move the final force unlink of the urbs from
usbduxsigma_free_usb_buffers() to usbduxsigma_detach().
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Initialize this variable with then_ai_urbs and n_ao_urbs in the
attach. Then usbduxsigma_alloc_usb_buffers() can just check the
variable to determine if the urb needs to be allocated.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'dac_commands' buffer is used to pass the ao channel list from
usbduxsigma_ao_cmd() to the urb callback, usbduxsigma_ao_urb_complete().
This buffer does not need to be allocated.

Change it into a simple array of the correct size. Rename the variable
to 'ao_chanlist' to clarify what it actually is.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Tidy up the error path by doing all the basic allocations then checking
for failures.

We have the comedi device now so fill in the urb->context.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This function never fails and currently returns '1' so that the
usbduxsigma_pwm_write() function returns to proper value (insn->n).

It's also called by usbduxsigma_pwm_config() which also returns the
'1' value. The proper return code for this function should be '0' to
indicate success.

Change the return type of the function to void and just have the
callers return the proper values.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This function never fails and currently returns '1' so that the
usbdux_pwm_write() function returns to proper value (insn->n).

It's also called by usbdux_pwm_config() which also returns the '1'
value. The proper return code for this function should be '0' to
indicate success.

Change the return type of the function to void and just have the
callers return the proper values.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename the local variable used for the private data pointer to 'devpriv'.

Tidy up the function a bit.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The pwm subdevice is not readable so this callback function is not
needed. Remove it.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove the sanity check of the private data. This function can only be
called if the private data was successfully allocated in the attach.

Tidy up the function a bit.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename the local variable used for the private data pointer to 'devpriv'.

Remove the sanity check of the private data. This function can only be
called if the private data was successfully allocated in the attach.

Tidy up the function to make it more concise.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename the local variable used for the private data pointer to 'devpriv'.

Fix the fx2delay calculation so it does not use floating point values.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The endpoint defines are each only used in one place and don't help
clarify the code. Remove the defines and just open code the values.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Generalize a helper function to replace for() loops in usbduxsigma_{ai,ao}_stop()
that call usb_kill_urb() to unlink all the urbs.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Generalize a helper function to replace usbduxsub_submit_{in,out}urbs().
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

wACKDuration
wRTSTransmitLen
byRTSServiceField
byRTSSignalField
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Always set to zero, so remove.
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changes:

* xillybus.h: __iomem added to struct xilly_endpoint -> registers to suppress
  "different address spaces" errors.

* xillybus_core.c: __user added as required for the same reason.

* The two member names of struct xilly_endpoint_hardware of the form
  sync_single_for_{cpu,device} were changed to something that won't look
  like the well-known functions.

* All *.c files: Variables and functions made static as required.
Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This variable was removed from the kernel a while ago.
Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Tidy up the multi-line comments to follow the CodingStyle.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

These parentheses are not necessary.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The allocation of the user range tables could fail. Make sure to check
for it.

Change the kmalloc()'s to kzalloc()'s to make sure the allocated range
tables are initialized to a known state.

Change the local variables so they can be used for both the ai and ao
range initialization and use shorter names to keep the lines < 80 chars.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

If the user does not specify the analog output range the private data
'user_ao_range_table' will not be allocated. The comedi core will detect
the NULL s->range_table and automatically set it to &range_unknown.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The function das16_reset() needs a valid dev->iobase. Since the iobase
is requested after the devpriv has been allocated in the attach, move
the das16_reset() call into the if (devpriv) and add a if (dev->iobase).

Also, move the release of the extra iobase into the if (devpriv) to
prevent an invalid memory access.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Change the MODULE_DESCRIPTION to something useful instead of the
generic "Comedi low-level driver".
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove the comments and add some whitespace to help readability.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Some of these messages are missing the terminating '\n' and most of them
have the quoted string split across lines.

Change all of them to dev_err() messages and fix the issues.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This function is only called during the board attach to determine the
pacer clock speed for the boards capable of bursting.

Absorb it into the caller and remove the attach noise.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Convert this printk message into a dev_err().
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This driver probes the board id during the attach to ensure that the
user is attempting to attach to the correct board type. Currently an
error message is output if the ids are not consistent but the attach
continues anyway.

Make the attach fail if the id bits do not match. Also, cleanup the
printk messages and use a dev_err() only if the probe fails.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Add the suffix _REG to all the register defines to better indicate
what they are.

Add a prefix to all the register bit defines to avoid namespace clashes
and to better indicate what they are.

Tidy up the whitesapce.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The 8255 digital i/o subdevice is only available on some of the boards
supported by this driver. Currently the 'i8255_offset' member of the
boardinfo is used to indicate that it is available.

To clarify this support, add a 'has_8255' bit-field flag to the boardinfo.
We can then use this flag to conditionally allocate the space for the
subdevice only when it is needed.

Also, make sure to check that subdev_8255_init() is successful and propagate
any errors.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

All the boards supported by this driver have an 8254 timer at offset
0x0c. Remove this unused information from the boardinfo.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Now that the boardinfo does not require any forward declarations,
move it near the definition of the struct.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The timer is only started by the ai subdevice in das16_cmd_exec()
and this function can only be called if the device is attached.

Remove the unnecessary sanity check.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

For aesthetics, move some of the functions to make the code easier
to follow.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Factor out the loop that waits for the ai conversion to complete.

Tidy up the function a bit.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Introduce a couple helper functions to enable and disable the
analog input interrupt/dma conversions.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This is just extra cruft. Remove it.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This variable is not used in the driver. Remove it.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>