Cesar Prados noted that if you boot w1 directly, not passing through
sockitowm, w1 was not working.  I always run sockitowm because I have
it in the lm32 code that starts every time I load the fpga.

This fixes the missing clock divisor.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This completes w1-eeproom.c with the complete API. The provided test
commands allow to read and write any size at any offset (the low-level
code manages cross-page reads and writes. The following examples
modify a pre-written eeprom, which I host in an external socket of
my SPEC card:

   wrc# w1r 0 4
   read(0x0, 4): result = 4
   offset    0 (0x000): 240 (0xf0)
   offset    1 (0x001): 241 (0xf1)
   offset    2 (0x002): 242 (0xf2)
   offset    3 (0x003): 243 (0xf3)

   wrc# w1w 0 40 41 42
   offset    0 (0x000):  40 (0x28)
   offset    1 (0x001):  41 (0x29)
   offset    2 (0x002):  42 (0x2a)
   write(0x0, 3): result = 3

   wrc# w1r 0 4
   read(0x0, 4): result = 4
   offset    0 (0x000):  40 (0x28)
   offset    1 (0x001):  41 (0x29)
   offset    2 (0x002):  42 (0x2a)
   offset    3 (0x003): 243 (0xf3)

   wrc# w1r 30 4
   read(0x1e, 4): result = 4
   offset   30 (0x01e):  69 (0x45)
   offset   31 (0x01f):  70 (0x46)
   offset   32 (0x020):  71 (0x47)
   offset   33 (0x021):  72 (0x48)

   wrc# w1w 30 11 22 33 44 55 66
   offset   30 (0x01e):  11 (0x0b)
   offset   31 (0x01f):  22 (0x16)
   offset   32 (0x020):  33 (0x21)
   offset   33 (0x021):  44 (0x2c)
   offset   34 (0x022):  55 (0x37)
   offset   35 (0x023):  66 (0x42)
   write(0x1e, 6): result = 6

   wrc# w1r 30 6
   read(0x1e, 6): result = 6
   offset   30 (0x01e):  11 (0x0b)
   offset   31 (0x01f):  22 (0x16)
   offset   32 (0x020):  33 (0x21)
   offset   33 (0x021):  44 (0x2c)
   offset   34 (0x022):  55 (0x37)
   offset   35 (0x023):  66 (0x42)
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

Signed-off-by: Aurelio Colosimo <aurelio@aureliocolosimo.it>

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

With this commit, when you select both the hardware and software uart,
ppsi diagnostics (controlled  by the "verbose" command) got to
the software uart, not to the hardware one. In this way you can run
the shell in a clean tty.

Unfortunately, frame dumps still reach the hardware uart, if frame
verbosity is greater than 1.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

Actually, I tried to move cmd_verbose.c to ppsi, and avoid
this dependency, but our use of --gc-sections would prevent
the code to be picked into wrc.o.

Hopefully, we won't need to touch this command any more.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This merges a number of changes related to a major ppsi update,
commits by me and Aurelio Colosimo.  This last wrpc-sw commit builds,
but previous ones do not. Still, I chose to commit separately each
fix for documenting purposes, with the maintainer's approval.

The new ppsi does not sync two spec cards, because I broke it during
my development session of April 2013, but lack of even build-test led
wrpc-sw master to not even build with its ppsi submodule.

This al least brings back build-testing to a sane state, and allows us
to concentrate on making it work and benefit from the new features on
WRPC-based cards.

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

We need to link ppsi/proto-standard/libstd.a explicitly, but we also
need a rule to generate it, or parallel compilation fails saying it
has no rule. The makefile depended on build order (my fault), as
building ppsi/ppsi.o, with its rule), also built libstd.a

BTW, this removes ppsi/arch-wrpc/libarch.a from the dependencies,
because it is already linked in making ppsi/ppsi.o (otoh the need for
an explicit mention of libstd.a can't easily be solved in ppsi
itself, so it remains here).
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

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

In recent ppsi code, proto-ext-whiterabbit is not replicating
proto-standard any more, but rather it offers hooks for the additional
code it needs.  Thus, we need to link the library for standard PTP as
well, to pick some of the states for the WR state machine.
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 source file is now part of ppsi itself, so it is removed from
wrpc-sw. The move helps improving cleaning up and maintaing ppsi
without committing to the supermodule each time something internal changes.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This commit updates ppsi with the latest development, but does not
build if CONFIG_PPSI is selected. This whole branch is devoted
to a smooth merge of ppsi changes, so every wrpc-sw modification that
depends on ppsi is commited and explained separately.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This commits affect several files, to be able to support both uart and
uart_sw at the same time. My aim is sending verbose ppsi diagnostics
to the uart_sw while retaining the real uart (or vuart) for the
interactive user and wrpc status gui.

To this aim, I use weak functions to avoid hairy ifdef around the code.
Now both hw and sw uarts are initialized, and the missing init just
does nothing. Also, uart_sw_write_string is always defined, and
it is an alias of uart_write_string if no uart_sw is built.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Sometimes a printf happens in the code, and it eventually prevents the
wrc to link because it pick up the whole of libc.  While ptp-noposix
already has a definition of printf to mprintf, we don't have the
same for ppsi, so let's add it.

Currently, rxts_calibrator.c includes calls to printf, and didn't
build with CONFIG_PPSI until this commit.
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 introduced an "advanced" menu, only for developers, and most
questions are grouped in there, because we cannot support users
playing with experimental features or work in progress (like the new
onewire code, or the new ptp daemon) until they are the officially
supported option.

The commit also removes changes in stack size, because the new
calibration code by tom is not stack-hungry like the previous one:
wrnic_defconfig is now the same as spec_defconfig and ppsi doesn't
select a bigger stack any more.

Finally, ppsi_defconfig replaces spec_ppsi_defconfig, since we only
support the "spec" hardware configuration (other implementations we know
of are still very much spec-alike).

Unfortunately, this spits two warnings:

    Kconfig:25:warning: defaults for choice values not supported
    Kconfig:29:warning: defaults for choice values not supported

even if such defaults actually works. The next commit removes them.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Split a rule into two rules, to make make happy.

In previous versions of make it was acceptable to list one or more
explicit targets followed by one or more pattern targets in the same
rule and it worked "as expected". However, this was not documented as
acceptable and if you listed any explicit targets AFTER the pattern
targets, the entire rule would be mis-parsed. This release removes
this ability completely: make will generate an error message if you
mix explicit and pattern targets in the same rule. You must split
these rules into two rules: one for the pattern and one for the
explicit targets. The Linux kernel source has already been modified in
this way (in newer kernels).

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

This also adds three shell commands, for testing:

     w1w PAGE TEXT
     w1r PAGE
     w1test

The w1test command writes to 80 pages and reads them back (warning: it
erases the whole device).

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

call t24p calibration every time PTP is set to Slave mode, or read the value from EEPROM when Master mode

Conflicts:
	dev/dev.mk

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

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

This adds the build choice about whether to use the older sockitowm or
the new "w1" implementation, which supports all three thermometers
but is still missing eeprom support (the API is defined, though, see
include/w1.h

This also adds a shell command, called "w1", which rescans the bus and
tries to read the temperature from all devices. In the example below
the third device is an eeprom, so it return an error for temperature
read-out.

   wrc# w1
   W1: 68000801dce56910
   W1: f70000001eda8242
   W1: 5f00000040e50143
   device 0: 68000801dce56910
   temp: 34.7500
   device 1: f70000001eda8242
   temp: 32.5000
   device 2: 5f00000040e50143
   temp: -32768.0000

The code is properly integrated in wrpc-sw, and the "first" thermometer is
used to build the MAC address of the device, like it used to be with
sockitowm.

The binary built with w1 is 3kB smaller than what is build with sockitowm.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>