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

zio_cset_is_self_timed() only returned true for input channels.  This
is needed, but the name is wrong: use zio_cset_early_arm() instead.

An output channel can be self-timed as well, so trigger_data_done
check the flag directly, not early_arm(). Every self-timed cset
must be rearmed when the trigger fired.

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

With the new self-timed output devices, the last raw_io() when the
buffer is empty would fail, so we need to mark the trigger as
not-armed, in order for output to restart when the next push happens.

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>

The ms-phase attribute pairs with ms-period. In this way you can have
a fine control about when the I/O happens.  Please note that the epoch
of the time base is unphatomable (it is related to boot time). So you
can manage offsets between csets, but not absolute stamps. For that
please use the HRT trigger.

Every trigger instance starts at phase 0 (i.e., the trigger is armed
at a multiple of its period). If you later change the period, no phase
calculation is performed, so that the new period will be as requested
(to allow for slowly-changing periods.  If, however, you change the
phase, the trigger is re-armed with that absolute phase, to allow for
slowly-changing phase.  If you change the period and then you change
the phase, you'll experience an unexpected period during the transition.

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

Federico Vaga authored 12 years ago and Alessandro Rubini committed 12 years ago

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>

This merges a series of commits that introduce self-timed devices
(exemplified by a tdc driver), zero-size-block for input channels and
an overall better and simplified locking policy for csets and
triggers.

When the current buffer is changed, there should be no
users on the channels (otherwise, either a user block
is being produced/consumed or there's somebody sleeping
on the wq of the buffer instance).

This prevents changing the buffer if in use, and prevents
users from opening a channel if the buffer is being changed.
We use the "disabled" flag for the buffer, not used so far.

We noticed this while working on buffer changes (see previous
commit). I'm unable to backport this to the proper place in the
history because it relies on the new locking introduced by this branch.

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

I'm unable to backport this commit to the proper place (too many
conflicts). Anyways, this fixes an issue related to changing the
buffer type (nothing bad happens if you don't change buffer type).

The bug is exposed by the self-timed csets: when a buffer type is
being changed, the trigger may be armed. This means that a block is
being used by the device, but the hosting buffer instance is being
destroyed.

This commit aborts a trigger and keeps it disabled while the buffer
is being changed. Re-enabling later, unless it was already disabled.

Even without self-timed support there was a tiny window for the race
to appear. Now it's fixed forever.

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

If the trigger is enabled on creation, it may fire before the cset is
ready.  This happened to me with "modprobe zio-zero trigger=irq".
For this reason it must start disabled. Also, the create
method MUST set ti->cset before activating the trigger, because the
cset lock is needed to access the enable/disable flag.

The commit adds a WARN() if the field is not set after creation, then
it leaves it zeroed to ensure the system will crash, instead of having
a subtle race.

Also, add a pair of FIXME notes about the need to allow changing
a trigger type while leaving the new one disabled.

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

This is a bug fix for overlooks in this same series of commits. I
won't fix each lock when it is introduced before proposing for master,
because it takes too much time.

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

Now we can have zero-size input blocks, as TDC is real. zio-dump
should not complain when reading zero bytes, if zero were expected.

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>

This has two csets, with one channel each. The first returns timestamps
in the data area, the second channel returns zero-sized blocks, with
the timestamp in the control.

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

These changes allow the driver to work with a PCI interrupt, a role
that was never used earlier. The commit fixes a warning on removal
when if libgpio is not present in the host kernel; it also tries to
request the shared interrrupt with various edge-trigger bits, until
one succeeds. This allows to use both edges for owned GPIO interrupts
and fall back to no specific requirement when used for a PCI
interrupt.

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

This commit allows zero-size channels to exist. The feature is used
for TDC and DTC (time to digital converters and the opposite).  This
simple change in the core allows to implement TDC drivers that only
return the control (which includes a time stamp) and no data.

For DTC to work we need some modification to chardev.c, which are not
there yet.

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>

This is important now that we have self-timed devices: if any trigger
parameter is changed (e.g., nsamples), we need to abort the armed
trigger and rearm with new parameters.  If the trigger is not armed
when changing parameters, nothing happens like before.

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>

This small patch supports self-timed input csets (for example, a TDC)
by ensuring the input trigger is always armed, waiting for the device.

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>

The flag is need so self-timed devices, like a TDC, can have their
trigger continuously armed (see next commit).

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

The function used to be called __zio_internal_data_done, in
heklpers.c.  Being used once only, it was expanded inline by the
compiler, so nothing changes for previous code.

However, if a trigger needs to implement its own data_done method, it
will most likely need to replicate this code, so it's better made
avaiable directly to everybody.  An alternative would be running this
code every time, before or after calling t_op->data_done, but we'd
better let each trigger choose the order of its own operations.

Actually, there is one difference: the input active block is set to NULL
after being stored to the current buffer, so the later implementation
of stop_io (used in zio-irq-tdc.c) won't expose a bug.

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

This is the first step of a series of changes aimed at a cleaner and
more flexible management of triggers, as well as a really transparent
"user" trigger.

This commit does the following:

- zio_fire_trigger renamed to zio_arm_trigger: the trigger is
actually armed by software, and then it fires by hardware. This
distinction is especially important for the transparent trigger: when
devices have internal timing, the software trigger must arm it
immediately (as opposed to the dumb devices, where the software trigger
really causes I/O to happen).

- ZIO_TI_BUSY renamed to ZIO_TI_ARMED. Also, ZIO_TI_COMPLETING is
removed. The new, simplified policy is like this: the trigger is armed
by software (the trigger module), but completion is driven by hardware
(the device module).  There is no need for a COMPLETING flag, because
the ARMED flag is only cleared after data_done is over for all channels
in the cset.

- the cset spinlock is used to protect all changes of the ti flags.

- change_current_trigger completely revised and fixed to match new
conventions.

- trigger->abort now takes "ti" instead of "cset" as argument; it is
more natural do do so, and no current trigger implements abort so no
harm is done.

- the trigger->abort and trigger->change_status methods are now
always called while holding the cset spinlock. ZIO core calls abort
when changing the current trigger type, to ensure no pending blocks
are there.

- zio_trigger_abort is renamed to zio_trigger_abort_disable, with
an additional argument to state whether the trigger must be atomically
disabled after the abort is over. This avoids a race condition on
trigger removal. It returns the previous "disabled" bit, to be used
when changing a buffer type while preserving trigger status.

- minor unrelated improvements in error management in objects.c.

- documentation update to match the new locking, and a few typos fixed.

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

The merge-data attribute tells the buffer to try a merge of a
newly-stored block with the previous one. When this is possible, the
buffer updates the previous block (nsamples, datalen) and releases the
new control immediately. This allows saving some memory without
breaking the ZIO data model. The feature only makes sense for input,
but there is no policy in place to enforce that.

When user space reads this new block, it will get attributes (and
timestamp) for the first data items, but all available items are
returned in this bigger block.

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

Federico Vaga authored 12 years ago and Alessandro Rubini committed 12 years ago

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

Federico Vaga authored 12 years ago and Alessandro Rubini committed 12 years ago

On load nsamples is 0 and this is wrong. It must be initialized with
the value in ti->nsamples. Otherwise if you read the current_control
on sturtup you will read 0 instead of the real value

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

Michael Auchter authored 12 years ago and Alessandro Rubini committed 12 years ago

This patch fixes a few typos in the zio manual.

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

Federico Vaga authored 12 years ago and Alessandro Rubini committed 12 years ago

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

Federico Vaga authored 12 years ago and Alessandro Rubini committed 12 years ago

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

pdftex has the very bad habit of picking png first, so we can't
have png images or the pdf version will be horrible. Thus, let's use
gif for the html (which is not picked by pdftex) and a .pdf generated
by our encapsulated postscript sources.

This adds the needed makefile and removes the previous pdf images,
which are regenerated in the next commit. We are also adding new
images where, including the logo used in the cover.

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