Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

The helper zio_trigger_abort() is introduced; it can be used to abort a
data transfer.

The bit flag ZTI_COMPLETING is introduced: if high ZIO is completing the
transfert operation by invokng the data_done() function; if low the
transfer is running or the trigger is waiting for event. If abort() is
called when this glag is high, the functions does nothing.

The helper zio_trigger_data_done() is introduced; insted of call the
data_done() function with cset->ti->t_op->data_done() and manually set
the ZTI_COMPLETING flag, this hellper must be used. If the trigger
operation data_done is not set, then use the __zio_internal_data_done

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

If default_(zbuf|trig) is not empty then ZIO use this as first choice
for the buffer|trigger; otherwise it tries to use the
preferred_(buffer|trigger) defined at zio_device and valid for any
cset. The last choice is the ZIO default.


Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

The core should ensure its users that create calls are not reentrant.
We use the locks that already exist in the buffer type and the trigger
type.

Then, we init both ti->lock and bi->lock, so our users can rely on
them (see next patch).

Related to locks: fix a comment in zio.h

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

This patch introduces zio_driver and zio_device_id structure,
which are used to define a zio's driver. ZIO uses these
structures to handle the match between devices and drivers
connected on the zio bus.

To make ZIO work on the Linux bus mechanism, ZIO creates two
zio_device structures. The first zio_device is allocated and it is
registered by the module which wants to declare the existence
of a new ZIO device. This first zio_device is only used by
ZIO to match the device with a driver. When the first zio
device matchs with a driver, ZIO creates a new zio device child
of the first one. The new device is filled with all the device
information and sysfs attributes. This second zio device
represents the real device to use; ZIO passes this zio device
as argument to the driver probe function, so the driver can
use the real zio device for its operations.
From the user space point of view, both ZIO devices are
available at /sys/bus/zio/devices/; the first zio device has
the prefix "hwdev-", the second zio device (the real one) has
the prefix "zio-"

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

cset and bi use a different set of sysfs default attributes, so now they
use their own device_type with their default attributes

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

remove the folder where zio lists all the available triggers, buffers
and devices. Now you can find devices at /sys/bus/zio/devices/, and
retrive the list of the available triggers and buffers from:
/sys/bus/zio/available_buffers
/sys/bus/zio/available_triggers

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Now all the zio objects are handled with the device structure

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

ad7887 and ad7888 are now handled by the same driver.
The driver requires data only for enabled channels

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Every device must has the zio attribute "resolution-bits" on any level of
the device hierarchy. Internally, the preferred name for fields
and variables is "nbits".

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

The current control used to be per-trigger (i.e., per-cset).
However, some parameters are per-channel (like the gain),
so each channel now has its own current control structure.

Also, zio-dump now shows device and trigger attributes, both standard
and
extended ones.

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

If a zio object declares a set of read only attributes and it doesn't
need s_op->info_get(), it can avoid the s_op assignment, so zattr_show
now checks if s_op is defined before calling it.

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

When disabling a trigger, ZIO invokes t_op->abort() to stop acquisition.
If the abort function invokes data_done() to return partial
blocks, the ZTI_BUSY flag is correctly cleared. But, if the abort
function frees all active blocks the flag remains set.

With this patch, always clear the ZTI_BUSY flag when the trigger is
disabled so it is correct if/when the trigger is re-enabled.

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

The bug was evident in zio_change_current_buffer. When you change the
buffer, zio registers a new bi, but the old __bi_register() overwrites
the current bi in use by the channel; so, when zio tries to unregister
the old bi, there is not any reference to it.

Now you must assign a bi to a channel after its registration. For
simmetry and to prevent future bugs, also ti assignment must be done
after __ti_registration().

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

Federico Vaga authored 13 years ago and Alessandro Rubini committed 13 years ago

To fix, set n_std_attr only if some std attributes are defined

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>

We always said we need analogue, digital and TDC/DTC support, so
add a type name for time converters. Meanwhile, reserve one bit
more so we may have up to 8 types -- but sure we don't want to state
what kind of physical measure does the analogue value represent.

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

When a device has several input or output csets, we usually want
a different data-transfer function for each of them. Using a function
specific to each cset simplifies things a little.

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

This patch adds the dev_id field in the control structure, to
identify different devices driven by the same driver, when the bus
abstraction will be completed.

Fields are also reordered: now hostid, devname, dev_id, cset_i and
chan_i are consecutive; thus, they can defined as a sockaddr_zio
structure for the forthcoming PF_ZIO infrastructure, without changing
the control structure again.

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Acked-by: Federico Vaga <federico.vaga@gmail.com>

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>

bugfix: add zio_device_put on error

Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
Acked-by: Alessandro Rubini <rubini@gnudd.com>