Changes also in wr_pstats.
Wr_nic now calls callback function in wr_pstats to get counters
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

By writing anything to a port file, you zero the counters for that
port. By writing to "info" or "description" you zero counters for
all ports.

Zeroing is performed by copying the current values to a new array,
and always subytracting that array to the current counter values.

No zeroing is performed for network statistics, that still reads
the "real" counters, featuring an overflow at 4G but no zeroing ever.

To simplify a little the code, I turned the one-dimensional array
with index arithmetics into a two-dimensional array, so the compiler
makes the math for us, transparently.

NOTE: the spinlock must be audited, we are not safe yet, not only
because of this patch.
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: Adam Wujek <adam.wujek@cern.ch>

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

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>

Also, run "wr_date get" at boot once, so it parses leap-seconds.list
and configures the kernel for the proper tai_offset.

With the previous commit it worked with "wr_date get tohost",
becase after the host is correctly in 2014 or so, all leap seconds are
in the past.  But if we are not synced, the host is in 1970 and
no leap second has already happened.

If this code is unchanged in 2018, we'd better be off by 2 seconds
than 37, so if linux reports a date earlier than 2014, use 2014 and
fix tai_offset at 35.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This merges a single commit, that I published some time ago for
our users, so they could set system time from WR time.

This is now going to happen automatically, in the next commits, but
this new wr_date binary is still useful to read /etc/leap-seconds.list
and pass the information to the kernel.

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

Long ago we used to refer to installed shared libraries and headers,
during further steps of user-space compilation.  We use static libraries
since long ago, and we have all -I pointing to ../ within the source
tree.  Get rid of unneeded installation of headers and extra -I
in the command lines.

Meanwhile, make all sub-makefile a little more similar one another
(this is trivial, mostly a side effect of the above).

The commit has no effect on build commands and results.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The files in userspace/include/at91 were not used.

Some were straight copies for linux headers, so including
<mach/at91_pio.h> instead of <at91/at91_pio.h> makes our copy
irrelevant.  This means both the hal and rtud must include from
$LINUX, but this was already a requirement for other userspace stuff.
The Makefiles are now fixed to do that.

at91sam9g45.h was only used for AT91C_BASE_SYS, which is no more.
This is trivially 4k-before-the-end, to mmap the end page.  We were
already defining AT91C_BASE_SYS_RAW to the exact same value (only,
AT91C_BASE_SYS was a pointer to structure), so the header could go as
well.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The output is now very verbose, but with grep and similar things we
can put it to good use. Besides, seeing all data items allows to audit
field use.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This version exports all status information in shared memory. This
allows a simplified wr_mon, and some diagnostics.  This is version 1
of the data structures, but we'll soon change them after some audit
to remove redundant, unused and misleading data items (or add the servo
structures, that are still missing).
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

In ppsi we allocate a big array of pp_instance items. And we want
to keep them all in shared memory.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

Dump of unknown memory regions was wrong in the "ascii" column.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

Use of attribute like this would have pointed to the previous bugfix
years ago.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

a long-long item wants llabs(), not abs().
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This branch is unifying the wr-nic driver in wr-swithc-sw and the
one in spec-sw.  The gateware cell is the same, so we are aiming
at a single driver, maybe SDB-driven in the future.

A few differences still remain, but they are very minor details, and
we'll fix them in further commits, without a specific branch.

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

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

Add posibility to setup MAC address at load time for SPEC.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

In kernel this function returns 1/true on success!
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

- Remove only for write in #72557df
- NetworkManger (Unbuntu LTS 12.04) insists on accessing to the phy
when it is used from LM32 and flood the dmesg so we remove the warning.

NetworkManager tries to access the phy, so this WARN_ON is
pretty common if the daemon is running. Now we know who is
misbehaving, and the driver can keep silent (disclaimer: this
opinionated message is by Alessandro :)

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

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

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

commit inspired by commit from spec-sw repo:
kernel/wr_nic: fix rx timestamps
c246c3ce36ac908ee3aada7db1f293bd12e42346

(commit edited for checkpatch by AR)
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

The calibration procedures are defined but never called. They
were used in wrs-V2 hardware, but callers disappeared in January 2012.
Better remove them than let them rust forever.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>