I'm confused by TAI/UTC: let's wr_date make it clear that it is
reporting TAI.  Nobody parses this output so far: the web interface
uses it, but it prints back the whole output.

So this change is safe.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Every other place calls the file wr_date.conf.

userspace/tools/wr_date.c will open the config file (it does not,
currently) but the name is different.

So we fix the executable and leave the rest as it is, since it works:
documentation matches S70wr_date and use of .config for build-time settings.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

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

Thanks to Cesar Prados for finding the problem
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>

This picks a new ppsi, which is needed for several reasons. We have
been using it for a while now, so it's a safe upgrade.  Being ppsi a
Kconfig thing itself, we now need to make wrs_defconfig first.

Also, the new ppsi.h defines INST which is pre-defined by net-snmp, so
include ieee1588_types.h alone, as we only need ClockIdentity by now.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

On higher storm of traffic, in the time between scheduling the tasklet and
executing it, I was loosing cntrs overflow information already (was overwritten
by all-the-time-incrementing counters in HDL).

File provides information about pstats version, number of counters per word
stored in memory and number of counters per each port.

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

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

We don't need a ppsi_defconfig any more, as it is now the default.
Also, use a full defconfig, so it's explicit what the options are.
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>

This fills the information by parsing the output of "wr_mon -g"
and "wr_mon -p".

Parsing is data-struct driven, and the strings must match the one
printed by wr_mon. If they don't a log message is generated:

   18:02:54 192 snmpd[1468]: wrs_ppsi_parse_line: can't find key "seource:"

To test, you can export WRS_SNMP_MON_FNAME and WRS_SNMP_MON_FNAME_P to
name two text files that can be opened instead of a pipe from wr_mon.

A problem with COUNTER64 remains, but it is a bug in netsnmp. See next
commit.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is needed if for example it is called in a pipe that is then closed.

meanwhile, remove trailing white space.
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>

Version items are now fake strings, as I still miss the code
that reads if from the device.  But they work, and the mib is
updated:

   tornado% snmpwalk -c public -v 2c -m +`pwd`/WR-SWITCH-MIB.txt wrs \
            .1.3.6.1.4.1.96.100.4
   WR-SWITCH-MIB::wrsVersionSw.0 = STRING:  fake-v4.0-rc1
   WR-SWITCH-MIB::wrsVersionGw1.0 = STRING: fake-7cce708
   WR-SWITCH-MIB::wrsVersionGw2.0 = STRING: fake-5118070
   WR-SWITCH-MIB::wrsVersionGw3.0 = STRING: fake-7efeb16
   WR-SWITCH-MIB::wrsVersionHw1.0 = STRING: fake-3.30
   WR-SWITCH-MIB::wrsVersionHw2.0 = STRING: fake-LX240T
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

thanks to greg for a 1-byte fix I squashed here
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

This is a squash-up of the whole netsnmp-pain branch, that I kept
on the repository, so to document the steps.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

I want to keep this scalar thing in place as a minimal working thing.
However, to rename an object I need to redo everything from scratch.
No comment on the quality of this all.
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>

This series of commits changes completely the installation and runtime,
moving to UBI.  I used it several times, and there are still minor problems
related to flash access, but it is generally stable.  I plan to find
the flash bug and fix it in a separate commit. This is ready for master,
in my opinion.

Note: a number of intermediate commits won't work, but I chose to split
the changes according to the package (barebox, kernel, userspace, ...)

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