This removes the register numbers from comments, and rewords
comments in general to match the fact that registers are now named.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The current code was reusing R10 and R11 for a different role.
However, we have several free registers, so this allocates two new
more, by giving them a meaningful name.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Uses of R20 and R12 are temporaries, and very local. Thus,
we can reuse R14, already blessed R_TMP.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The existing code used R6 both for ARP and for ETHERBONE.  Allocate a
new register, since we have plenty. This clarifies the code (by using
proper symbolic name) but changes the generated binaries.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This commit is not changing the generated code at all (the
bits generated are exactly the same as in the previous commit),
in all three cases of #ifdef.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

By naming registers the code is much more readable in my opinion.
And we avoid, in future modifications, to mistake a frame offset for
a register.

The enumeration of names starts from 1024 to deny use of simple
numbers. Also, the enumeration is verified to be exactly 32 names,
by raising a build error if not.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

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

"act accordingly" means pre-including ppsi/...autoconf.h within
CFLAGS, or we miss CONFIG_MAX_VLANS_PER_PORT, used in defining the
ppsi local data structure.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

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

The current default is a multicast address. We'd better use a
locally-assigned number instead. We keep the pattern of a very-visible
thing, so if it appears in some sniffer or ARP table you notice.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Because the golden bitstream for SPEC has syscon at offset 0x0.

E.g. Kintex-7

Based on Peter's work for km2net CLB board.

Reason: h_y was close to the DAC range, on some boards we were unable to lock
HPLL.
In addition to HPLL modifications, also the phase shifting had to be changed.
When offset clock has lower frequency than the ref clock, shifting ref
clock produces sampled clock which is shifted in another direction.

The function ptpd_netif_create_socket() still receives two arguments,
because it is being called by ppsi, and I'd better not change the API
in small steps in there.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

y

The function wrpc_get_port_state() still receives two arguments, because
it is being called by ppsi, and I'd better not change the API in small
steps in there.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The function are not there:

 * ptpd_netif_poll()
      it never existed

 * ptpd_netif_select()
      nobody called it, and it only returned 0

 * ptpd_netif_locking_enable()
 * ptpd_netif_locking_disable()
 * ptpd_netif_locking_poll()
 * ptpd_netif_adjust_counters()
 * ptpd_netif_adjust_phase()
 * ptpd_netif_adjust_in_progress()
 * ptpd_netif_enable_timing_output()
 * ptpd_netif_enable_phase_tracking()
 * ptpd_netif_read_calibration_data()
      they were part of ptp-noposix-wrappers.c, removed in January 2015.
      The features are now part of the wr_ops in ppsi.

 * ptpd_netif_get_ifName()
      same (ptp-noposix-wrappers.c), it returned a constant name, and
      ppsi does it differently because based on arch-wrs/arch-wrpc

 * ptpd_netif_calibrating_enable()
 * ptpd_netif_calibrating_disable()
 * ptpd_netif_extsrc_detection()
      same, but in addition they were already doing nothing when removed

 * ptpd_netif_calibrating_poll()
 * ptpd_netif_calibration_pattern_enable()
 * ptpd_netif_calibration_pattern_disable()
      same, but they wer not actually used since calibration was gone
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 time I added -Wmissing-prototypes and fixed accordingly. Again, this
used to be part of -Wall and I'm disgusted by it not being so any more.

What does "all" mean, then?
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

I added -Wstrict-prototypes, that used to be included in -Wall settings
but is not.  Actually, I think it should be an error to not specify an
argument list by now.  If any, -traditional is there for old timers.

Fortunately no bug was exposed by the missing prototypes.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Yesterday I was eventually fussed up by the need to "make clean" every
time. So here's a stupid depend rule, not perfect but better than
nothing.
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 adds two commands: devemem (to read and write registers/RAM)
and delays (to read and change the constant delays.  The commands are
only available for CONFIG_DEVELOPER builds.

The default delays were just wrong, they are valid of an older
gateware version.  Using 0, it is clearer to the user that the right
value is missing (if the SFP database is populated, the right value
applies anyways).
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

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