The original code was broken because we do use the dependencies for the
static analysis. However, it was silently failing. Removing the default
REPO_PARENT made the problem visible.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Resolve "fmc-fdelay-pulse by default does not set pulse width to 50%"

Closes #7

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!5

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Resolve "Update to latest dependency releases"

Closes #5

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!6

Resolve "migrate to common CI and build bitstreams"

Closes #14

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!3

Resolve "Incorrect string path for open_by_lun"

Closes #11

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!4

Except general-cores and wr-cores that are updated to latest working commit used in
other projects as well.

Closes #16

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

The path size was allowing only 2 digit LUNs. It never happened that we
have a LUN greater than 9 [0, 9]. However, the LUN is an arbitrary
number, any value is a good value. Therefore, increase the string size
to potentially contain a 32bit number (10 digits).
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Resolve "Kernel OOPS after few minutes of loading fine-delay svec module"

Closes #10

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!2

In the commit:
	commit 02ec43ce74e015eed509305e4faad78c9a8f828a ("kernel:
	spec_core: Update API setup_timer() to timer_setup()")

We added the newer protoype for 'fd_update_calibration()' along-with the
older one with conditional ditrectives.

Reflect the same changes in 'fine-delay.h' to avoid compilation error.
Signed-off-by: Vaibhav Gupta <vaibhav.gupta@cern.ch>

This reverts commit 25f66f51.

Turns out that even though Red-Hat backported "timer_setup()" on CS7, it is
not working well and corrupting the memory somewhere. Thus, we should
stick to the API provided by vanilla kernel.

Minor cleanup

Resolve "review vaibhav changes"

Closes #8

See merge request be-cem-edl/fec/hardware-modules/fmc-delay-1ns-8cha!1

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

It avoids ImageMakic errors when containers have stricter policy
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

As per tovalds/linux, the kernel version below 4.14 support
	fd_update_calibration(unsigned long arg)
and not
	fd_update_calibration(struct timer_list *arg)
But even though CS7 is 3.10 based, patches might
have been backported, and it supports the latest one.
Signed-off-by: Vaibhav Gupta <vaibhav.gupta@cern.ch>

Signed-off-by: Vaibhav Gupta <vaibhav.gupta@cern.ch>

Signed-off-by: Vaibhav Gupta <vaibhav.gupta@cern.ch>

This warning is treated as error by the compiler. Mark it with
"__attribute__((unused))".
Signed-off-by: Vaibhav Gupta <vaibhav.gupta@cern.ch>