Otherwise we have a race condition with wr_date setting initial time on the
switch.

in ppsi:
--update hal_shmem to version 7
--add tx and rx packet counters
--add error counters for SNMP
--increase ppsi shmem version to 6
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

remove waiting state, make waiting flag instead
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Add checking stack overflow and detecting reset to wrs. Code common for wrpc
and wrs is placed in system_checks.c and .h. Work is based on following
commits in wrc_main.c:
1704a668 main: check for stack overflow
7836c39e Optionally detect reset and print a stack trace

Additionally:
--update Kconfig to have a possibility to select CHECK_RESET also for switch,
  not only for node as before
--and _endram to ram-wrs.ld

When "stack overflow" occurs then message is printed once every second.
When reset occours then LM32 will print stack trace and reset once more.
NOTE: I couldn't get spll to work after reset, for sure problem is in
      ad9516 init function, but not only there.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

--increase spll_stats structure version to 2
--add to spll_stats structure:
  - commit_id
  - build_date
  - build_time
  - start_cnt (to be used later for start counter)
--move declaration of spll_stats structure from wrs_main.c to revision.c
--pass git version and author at compile time to revision.c
--create revision.h with build_* externs
--update version reporting at cpu start
--update version command
--include section .stats in section .data for node
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Do not report to monitor oscilations between state WR_WAIT_SYNC_IDLE and
TRACK_PHASE. Real problem will be fixed later.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Needed since ppsi moved to shmem and removed cur_servo_state structure from
servo.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

ppsi for wrs requres locks in shmem when shmem is changing. Since servo is
shared between wrs and wrpc fixes had to be implemented.
Since there is no shmem in wrpc, dummy function for locking had to be added.
As a side effect include/libwr/shmem.h was added.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

pass of header cleanup from wrs
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Done as a part of work to move hal to shm in wrs.

Done in this commit:
--use wrpc_get_port_state instead of halexp_get_port_state
--make wrpc_get_port_state fill hal_port_state structure in the way how HAL
  fills shmem.
--update lib/net.c to use struct hal_port_state instead of hexp_port_state_t
--update monitor/monitor_ppsi.c to use struct hal_port_state instead of
  hexp_port_state_t
--update ppsi to get needed changes
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Files needed to get rid of halexp_get_port_state.
Done during process of moving HAL to shm in wrs.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

It is not used anymore.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Update include/hal_exports.h to latest version
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This is useful for example when using a secondary onewire controller.

This reverts commit 7e2ee9b4.

You can't put a bitstream into the middle of a tiny EEPROM.
If you intend to build an sdbfs for a switch, make a new image.

It's a preparation for flash support, as there we cannot update
bytes/entries count without erasing entire flash sector.

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

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

- copyright header to local conventions
- fix a little white space
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

- Turn the copyright header to the local convention
- fix a little white space
- fix flash_sdb_check (must return error -- but it's unused)
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The optimization is the removal of the retval variable in
bbspi_transfer() and using the val parameter as a "shift register"
that is returned at the end of the function.
Signed-off-by: Theodor Stana <t.stana@cern.ch>

How-to guide for the tool (together with flash-write tool) can be found in
sdbfs.README.
Signed-off-by: Theodor Stana <t.stana@cern.ch>

SQUASH remove flash-read binary: 598d cdf68
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The issue was due to a bug in gpio_in, which was doing an AND with
the value of GPIO_SPI_MISO, which was set to the pin offset, instead
of what it was set in the include/syscon.h file.

When I copied the gpio functions from include/syscon.h, I overlooked
the fact that the pins for the gpio functions there are set using
wbgen macros that actually perform the shifting, and I simply copied
the functions and redefined the GPIO_SPI_MISO macros as offsets.

The issue has been fixed by redefining the gpio_in function as an
AND with the value (1 << PIN_NR). This fixes everything.
Signed-off-by: Theodor Stana <t.stana@cern.ch>