Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

To avoid version mismatch
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Avoid local copy of hal_minirpc.c. Anyway other (.h) files from
wr-switch-sw are needed to build ppsi...
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Avoid local copy of hal_exports.h. Anyway other (.h) files from
wr-switch-sw are needed to build ppsi...
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Files added to hal_headers are actually libwr headers.
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Catch up with PPSi changes.
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Avoid changing indent.
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

It is not checked in other parts of PPSi whether the pointer to
get_GM_lock_state is not NULL. To avoid crashes add dummy function.
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

It is not checked in other parts of PPSi whether the pointer to
enable_timing_output is not NULL. To avoid crashes add dummy function.
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Simplify an include path
Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Ben 211018 relicensing

See merge request !3

And also rename the COPYING files with GPLv2 and LGPLv2.1 suffix to
avoid any confusion.

After a discussion with CERN, it has been agreed to allow a relicensing
of mini-rpc source code to LGPL v2.1 in order to fix LGPL license for
ppsi when compiled for arch-wrs (the GPL of mini-rpc was breaking the
intention to realase PPSi as LGPL for arch-wrs).

The source code of mini-rpc is fully under CERN copyright with some
fixes from Seven Solutions, and all parties agree for this relicensing.

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Like:
--prio1
--prio2
--domain (domain number)
--class (clock class)
--accuracy (clock accuracy)
--allan (clock allan variance)
--tsource (time source)
Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Such change simplifies the implementation of a command changing priority1,
priority2 and domain number in runtime.
Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Used by WRPC
Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Fix the stuck in UNCALIBRAT/IDLE after e.g. accuracy changed
Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

This is not the final solution. With this fix this error does not occur
any more, but WRS/WRPC does not sync. It hits the bug #36Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Before this commit activePeer was filled when the siwtch was in master
mode or when the external port configuration was active.
This commit adds storing the MAC of each foreign master, the MAC of
the chosen foreign master is copied to activePeer.

IMHO: activePeer (or peer in general) should not be the MAC but
the portIdentity of a port on the master.
Signed-off-by: Adam Wujek <adam.wujek@creotech.pl>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>

Signed-off-by: Adam Wujek <dev_public@wujek.eu>