jhash changed from 2.6.24.7-rt27, so we have to lift it from
a later kernel

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is below spec, and the driver must ignore it

Amending 289428fe, delay_h is not taken into account, given that
the specification excludes delays > 200s.

Conflicts:
	kernel/fd-zio.c

This patch addresses the performance problem reported
by Christophe Chanavat (a single call to fdelay_config_pulse
takes 2.1ms to complete).

The time is spent in driving the GPIO enable line that
controls the output stage decoupling. This is done via SPI,
and each spi_xfer takes 420us, which add up to the reported
delay.

The fix leaves the enable pins always on, so the output relay
is not used to control whether the outputs are enabled/disabled;
it is the DCR enable bit who takes this role instead.

This results in an average 8us time for output delay configuration,
3us for output disable.

Conflicts:
	kernel/fd-zio.c

This is far more informative about what specific mezzanine
originates an error, e.g., low-level traces like

gpio_writel_with_retry: succeded after 66 retries
gpio_writel_with_retry: succeded after 65 retries
gpio_writel_with_retry: succeded after 97 retries
...

coming from nameless devices require identification of the originator.

This is below spec, and the driver must ignore it

If fdelay is programmed with -1 or some other random
value for the repetition rate, a drifting delayed pulse
will appear. Only rep = 1 makes sense

This avoids locking the board by programming a bogus delay
plus disable repeatedly

For traceability purposes, we include commit descriptions
as MODULE_VERSION information in drivers that are deployed
in CERN control system.

Ideally, the commit description should be added to all the
binary blobs installed, but the kernel module is critical
in this connection, as it is crucial to trace back quickly
the whereabouts of installed .ko sources.

These additions were overlooked in b7b800ae

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

There was an error in the overflow check of coarse values.
This replaces "> 125*1000*100" with ">= 125*1000*1000"
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

These amounts are *signed*, so we add this kludgy duplication
of code to handle the case of addition of a signed quantity
in fd-irq.c

The arithmetic code was suggested by Alessandro Rubini,
replacing a former, buggy and kludgy implementation by
David Cobas.
Reported-by: Alessandro Rubini <ru@gnudd.com>
Reported-by: Tomasz Wlostowski <twlostow@cern.ch>

This ugly hack (hardcoding the base address of slot 1 fdelay
core) is needed in the SVEC, as we have yet no means to discover
that core's base address via SDB (and will never have, probably).
The slot number of the core would probably better be taken from
a register in the core itself.

Reported-by: Tomasz Wlostowski

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

lib/fdelay-pulse-tom: dump start/end/delta timestamps. Default pulse width to 250 ns if not specified
Signed-off-by: Luis Fernando Ruiz Gago <lfruiz@cern.ch>

Signed-off-by: Luis Fernando Ruiz Gago <lfruiz@cern.ch>

The new calibration is now there, but some details are still missing,
like using the DAC value for the VCXO and validating against the
installed gateware.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is a massive change, spread across several files.
     - the calibration structure is changed
     - it doesn't live at offset 0x1800 any more
     - it must beloaded by SDB structures (not there yet)
     - all calibration users are fixed
     - calibration.c is a new file dealing with it
     - i2c.c has some material less, because of calibration.c
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The SDB library is an exact copy of the sdbfs/lib directory of repo:
   git://ohwr.org/hdl-core-lib/fpga-config-space.git
at version "sdb-v1.1" (commit de2c98f).

It actually is the same code that is already in fmc-bus/sdb-lib, but
until I finalize eeprom SDB support within fmc-bus I need a copy here.
This directory will disappear when fmc-bus is completed in that respect,
but we need a release with SDB in the eeprom now for fine-delay.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This old FIXME is now fixed. Unfortunately the calculation involved is
very boring, given the range of values we deal with.

With this in place, we can now generate a 10MHz square wave and
simulate the inputs for a WR grandmaster.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>