By doing this we force to read again the FPGA register inside the loop
     and the compiler can not skip this loop thinking he is reading the
same value (that was not modified by him).

In some case the softpll was not properly set (0 => unknown) when
checking for WRS mode.
We have force to wait at max 20s for shmem to be properly initiated.

By doing this we ensure to have started locking the GM mode (wr_date
check for soft pll shmem). And we have properly set FPGA time and leap
second before starting ppsi

Fix TAI offset into kernel was done withing get function but this has
been removed in commit #91f4c8d. As it is still needed by slave, we
force its use using -n option. This will only fix tai offset into kernel
without applying ToD to WR/FPGA time.

The return offset is only display so that we can check its value if the
user send us back the log but it is not used in any other way.

- Check if we are in GM mode, if yes :
    - wait GM to be locked
    - Disable WR as kernel clock source
    - Increase retries for NTP
- Retries x time NTP until setting host time:
    - Set date from host to WR
    - Enable (back) WR as kernel clock source

- Calling twice this function does not adjust nanoseconds, the user need
to call adjns function in order to perform this.
- Move printf only after setting time to be more deterministic

Use gmtime instead of localtime (TAI should never used localtime)
Add verbose option to also display linux time

The output will be something like:

wr_date -v diff
TAI(HW)-UTC(SW): +36.602286
UTC(HW)-UTC(SW): -0.397714

In the previous version `wr_date get` would override kernel TAI time
(using local file) and this could cause some problems in case leap
second from /etc/leap-seconds.list was not the same than from PTP/WR
protocol.

We have 37s of leap second but we were showing without usec the
following which could be seen as a wrong leap second offset.

WR time (TAI):     2019-03-15 05:40:48.006697
Switch time (UTC): 2019-03-15 05:40:10.975551

localtime has also been replaced by gmtime (TAI should never be
converted using localtime in case TZ variable is set)

The previous fix didn't solved completelly this issue: Negative nano WR
time adjustement generate a big jump of the monotonic clock used in
PPSI.
The current fix solves this issue.

The WR time adjustment was not taken into account in the driver. The
side effect was a big jump sometime of the monotonic clock used in PPSI
for the calculation of the timeouts. The side effect was that many
timeouts were firing too early.
The driver registration take now into account the adjustment and also
the calculation of the offset in the read() call.

since v5.0.x the WRS needs more time. We prefer to increase the reboot
time in order to properly refresh the web page after a reboot.

Remove the stderr in the shell_execution to improve parsing without the
state of the ports properly.

According to php manual (http://php.net/manual/en/function.strpos.php)
strpos must be used the indentical comperator because strpos can return
0 if string found at position 0.

Now we use ajax+js to constantly refresh value of WR into the web
interface without needing to manually refresh the webpage by the user.
However this refresh flood the accesslog and this is a problem with the
small RAM available

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

bug was introduced in www: removed current setting
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

I saw that in the endpointconfiguration.php you relay a lot on an order of parameters. In other words the following line will be displayed correctly:
CONFIG_PORT01_PARAMS="name=wri1,proto=raw,tx=224295,rx=225959,role=slave,fiber=0"
but not when I swap an order of parameters:
CONFIG_PORT01_PARAMS="name=wri1,proto=raw,tx=224295,role=slave,rx=225959,fiber=0"
I'm asking you for that because there are three more new parameters that are optional:
      "ext" (extension) can be one of the following:
        - WR - use White Rabbit as an extension on this port (default if not present)
        - none - no extension on this port
      "dm" (delay mechanism) can be one of the following:
        - e2e - end to end (default if not present)
        - p2p - peer to peer
      "monitor" option to disable ("n") or enable ("y") triggering errors in SNMP on a port; "y" by default

NOW: it allows to change the order and dropdown menu in several parameters

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

And use it if exists
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Could you change it a little bit more? Can you make it to the way if there is no file /tmp/www_wrs_version then execute the command. Otherwise read the existing file

please remove the current setting from the first line. When the port is configured as slave the first and the second entry is slave.

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

can you change to have the dropdown list with modes in Endpoint Mode Configuration and in "Endpoint Configuration"

WR Status "UNDEFINED" when no connection to master, can you make just an empty cell?

caching wrs_version; commit: www: [BUG #1528] cache wrs_version; I would prefer this is done once for all sessions. i.e. if there is no file /tmp/www_wrs_version then execute the command. Otherwise just read content of this file

Fixed Calibration tables to write in dot-cong correctly

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

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

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

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

renamed:    www/endpointcalibration.php -> www/endpointconfiguration.php
renamed:    www/fibercalibration.php -> www/fiberconfiguration.php
renamed:    www/sfpcalibration.php -> www/sfpconfiguration.php
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

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

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