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>

patch taken from spec-sw repo:
kernel/wr_nic: major cleanup of tx timestamping
ab3f2f8d18e4f6ed744438314e5a78b8a509de74

(commit amended by alessandro to avoid #ifdef in the functions)
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This change adds warnings, but they will disappear after move to 3.14,
where %pa is supported.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This fixes an oops when programs access ethtool methods, especially
wrn_get_drvinfo. It can be reproduced by a simple run of "irqbalance",
which I never did in my development machine.
Reported-by: Grzegorz Daniluk <grzegorz.daniluk@gmail.com>
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

EOF

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>

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>

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

This branch adds WRS shared memory, and makes the HAL use it.
Clients (both locally with libwr and ppsi without libwr) now
access port status with shared memory.

We also have a shmem-dumping tool, that helped auditing the
data structure (but changes to them are not applied yet).

As a side effect libwr and the hal are passed throught Lindent,
but before any change, so you can still blame these changes.